Kelly B. Heaton
& The Coherence Research Collaboration
March 28, 2026
Abstract
Maxwell’s equations are governed by two independent constants: the propagation speed \(c\) and the vacuum impedance \(Z_0 = \sqrt{\mu_0/\varepsilon_0}\). The standard formulation of electrodynamics retains \(c\) and discards \(Z_0\), a compression that is exact for propagation questions but lossy for questions about how field energy partitions between the electric and magnetic modes. This paper restores \(Z_0\) as an explicit coordinate and derives its physical consequences without free parameters.
The geometric structure generated by the two vacuum constants is the Riemann sphere, with \(\varepsilon_0\) at the source pole and \(\mu_0\) at the sink. The logarithmic field between the poles has a unique self-similar attractor, the golden ratio \(\varphi\), selected by the Markov spectrum theorem. The irremovable defect at the source pole displaces the actual winding ratio from \(\varphi\) by exactly \(\alpha = Z_0/2R_K \approx 1/137\): the fine-structure constant, derived with zero free parameters. The orbital impedance \(Z_{\rm orb}(n) = Z_0\,n/\alpha\) follows from Coulomb’s law and the Biot–Savart law, with charge and radius cancelling exactly from the ratio \(|\mathbf{E}|/|\mathbf{H}|\).
The vacuum field potential \(\phi\) satisfies Laplace’s equation. Because \(E = E_0\,e^{-\phi}\) and \(\nu = E/h\), it follows that \(\log\nu = \mathrm{const} - \phi\), and therefore: \[\nabla^2(\log\nu) = 0.\] This is the Harmonic Light Law: the logarithm of photon frequency is a harmonic function of the vacuum geometry. Its unique spherically symmetric solution is the Thread Law \(\log_{10}\nu = \chi + \beta\kappa\) with \(\beta = \log_{10}\alpha\); its circuit projection is the I–V admittance \(dI/dV = -1/Z_{\rm orb}\). Two instruments — spectroscopic and electrical — read one equation. Minimum description length analysis across six independent spectral domains (atomic, solar, stellar, molecular) recovers \(\beta = \log_{10}\alpha\) with zero free parameters, confirming that photon frequencies in nature organize as a harmonic function.
The impedance-match condition \(Z_{\rm orb} = Z_0\) requires \(n = \alpha \approx 1/137\), satisfied by no positive integer. The field configuration this condition defines is simultaneously required by the geometry and forbidden by the closure condition of any bound state. The consequence is a structural gap in every ion’s transition spectrum at \(\Delta E = IE/4\), where \(IE\) is the ionization energy and nuclear charge cancels exactly. A direct survey of the NIST Atomic Spectra Database across 80 ions confirms a zero-observation interval at \(IE/4\) in 71 ions, with zero falsifications. Five sub-0.1 meV precision-tier ions return a mean corrected residual of \(-2.8 \pm 2.7\) ppm (SEM), recovering the electron rest mass from the location of a structural absence through \(m_e c^2 = 8G/\alpha^2\), where \(G\) is the measured gap center. Control analyses at \(IE/3\) and \(IE/5\) return \(\pm 250{,}000\) ppm, in exact agreement with the algebraic prediction for wrong-fraction evaluations. The same boundary appears at \(S_n/4\) in all four doubly magic nuclei surveyed and is absent in all 20 non-doubly-magic nuclides, with zero overlap between populations.
The fine-structure constant, Planck’s relation, the electron rest mass, and the Bohr radius are not independent inputs to this framework. They are consequences of a single geometric structure — the Riemann sphere generated by the two independent constants of the electromagnetic vacuum — locked together by exact algebraic identities and readable from a single structural absence in the spectrum. This is a geometric unification of the electromagnetic scale: more derived from less, with zero free parameters, zero new physical entities, and zero falsifications across 71 atomic ions and two independent physical scales.
============================================================
Maxwell’s equations are governed by two independent constants: the speed of light \(c\) fixes how fast the electromagnetic field propagates, and the vacuum impedance \(Z_0 = \sqrt{\mu_0/\varepsilon_0}\) fixes how the field energy partitions between the electric and magnetic modes. Neither determines the other. Both are required to fully characterize the electromagnetic vacuum; yet for over a century, one of them — \(Z_0\) — has been systematically set aside.
This was not an error in the domain of applicability: the compression of \((\varepsilon_0, \mu_0)\) into \(c\) alone is exact for questions about propagation speed, and propagation speed is what most of classical and quantum electrodynamics requires. But the compression is not exact for questions about how the field energy is partitioned — about impedance, about the ratio of electric to magnetic amplitude, about what happens at the boundary between a bound field and a free one. For those questions, \(Z_0\) is indispensable. And those questions, it turns out, are the questions that determine the fine-structure constant, the structure of atomic spectra, and the electron rest mass (Coherence Research Collaboration 2026; K. B. Heaton and Coherence Research Collaboration 2026c).
The present work began with an observation that we could not initially explain. In developing the Thread Frame coordinate system for atomic spectra (Kelly B. Heaton and Coherence Research Collaboration 2025b; K. B. Heaton and Coherence Research Collaboration 2026a), we found that photon frequencies across multiple atomic series organized on a logarithmic slope \(\beta = \log_{10}\alpha\) with a regularity that could not be accidental (Kelly B. Heaton and Coherence Research Collaboration 2025a). We did not know why. Separately, the derivation of orbital impedance produced a frequency-ratio relation involving \(\alpha\) and \(Z_0\) that appeared to belong to the same family of results. The two observations sat alongside each other without a common explanation. While completing the geometric derivation of the fine-structure constant from the Riemann sphere (Coherence Research Collaboration 2026), the connection became visible: both observations were readings of the same underlying equation: \(\nabla^2(\log\nu) = 0\). The vacuum field potential \(\phi\) satisfies Laplace’s equation; the energy satisfies \(E = E_0\,e^{-\phi}\); Planck’s relation gives \(\nu = E/h\); therefore \(\log\nu = \mathrm{const} - \phi\), and since \(\phi\) is harmonic, so is \(\log\nu\). The logarithm of photon frequency is the harmonic function of the vacuum geometry. The spectroscopic slope and the circuit relation were not two results; instead, they are two instruments reading one law, herein named the Harmonic Light Law. Independent confirmation came from minimum description length analysis across six spectral domains (Kelly B. Heaton and Coherence Research Collaboration 2025a) — atomic, solar, stellar, and molecular — which recovered \(\beta = \log_{10}\alpha\) with zero free parameters. A pattern became a law.
We did not modify Maxwell’s equations or contradict any experimentally established result. Neither did we add new particles, fields, forces, or dimensions. What we accomplished is in some ways more demanding: we identified the geometric structure that the equations already contain but do not make visible, traced its consequences without introducing free parameters, and found that those consequences include a new law — the Harmonic Light Law \(\nabla^2(\log\nu) = 0\) — and a precise spectroscopic prediction at \(IE/4\) confirmed across 71 atomic ions with zero falsifications. The fine-structure constant, Planck’s relation, the electron rest mass, and the Bohr radius are not independent inputs to this framework. They are consequences of its geometry, locked together by exact algebraic identities and readable from a single structural absence in the spectrum. The physics was always there. The coherence is new — and the result is, in a precise sense, a geometric unification of the electromagnetic scale.
The dominant tradition in modern theoretical physics rewards ontological novelty: new particles, new symmetries, new dimensions. This is not without reason — such discoveries have transformed our understanding of nature. But the accumulation of primitives is not the only form of progress. A theory that derives more consequences from fewer assumptions is strictly stronger, by any rigorous criterion of explanatory power, than one that requires more assumptions to derive less. The Standard Model requires nineteen free parameters. The framework developed in this series of papers requires none. That is not a stylistic difference; it is a structural one.
As we will demonstrate, a deeper understanding of the periodic table, the fine-structure constant, the electron mass, and the universality of atomic spectra does not require any new physical entities. The organizing principles follow from two primitives that were already present in Maxwell’s equations, once those equations are read with both of their independent constants restored. Simplification and coherence are not lesser achievements than new particles or forces.
This paper does not merely assert or prove, but teaches. It builds a reader’s model step by step, earning the right to each new concept before introducing it. The proofs are in the companion papers listed below. The purpose of this paper is to make those results intelligible — not just technically correct, but genuinely understood, and understood in sequence, so that each result is seen to follow from the one before it without remainder.
Our ideal reader has a solid background in physics, or knows how to read collaboratively with an AI interpreter — the method by which this entire body of work was produced. We assume familiarity with algebra, geometry, Maxwell’s equations, atomic spectra, and transmission line theory. We encourage professional skepticism: not hostility, but an unwillingness to be persuaded by enthusiasm alone. Every step in this paper is either a mathematical theorem, a definitional recognition, a derivation with no free parameters, or a direct empirical observation. Nothing is taken on faith, and nothing requires agreement with our interpretive framework to verify. The chain is: premise, derivation, consequence — repeated eight times, from the two vacuum primitives at the beginning to the electron rest mass at the end. We invite our reader to follow each step, notice if anything is missing, ask why each result could not be otherwise, and check the numbers.
Companion papers.
This paper culminates a series of works beginning in 2025:
The present paper is the teaching companion to papers 5 and 6: paper 5 proved the geometric foundation; paper 6 confirmed it empirically. This paper explains both. A reader who wants the proof without the teaching should read papers 5 and 6 directly. A reader who wants the teaching without the proof should read this paper and treat papers 5 and 6 as references.
============================================================
This paper is about the electromagnetic vacuum. We do not accept the premise of a spacetime container and study what matter does inside of it, but instead ask what the vacuum truly is — what generates it, what structure it necessarily has, and what that structure implies for every atom that it sustains.
We begin before Maxwell. Before measurement. Before the assignment of any numerical value to any physical constant. We begin with the only question that is genuinely prior to all of these: what is the minimum structure from which the electromagnetic vacuum necessarily follows?
The answer has three parts: two irreducible primitives and the relationship that their encounter generates. Understanding why three is the minimum — and why the structure takes exactly the form it does — is the task of this section. Everything that follows in this paper and our associated works (Coherence Research Collaboration 2026; K. B. Heaton and Coherence Research Collaboration 2026c) is consequence.
There are two kinds of thing that are logically prior to the electromagnetic vacuum. Not two kinds of particle, not two kinds of field, not two kinds of force. Two kinds of condition: the condition for boundary, and the condition for the absence of boundary.
The first condition is the capacity for demarcation: the property that makes it possible for something to be here rather than everywhere. This is the condition for a boundary to exist, i.e., for an inside to be distinguishable from an outside, for a form to be closed rather than open. We call this the structural primitive, and we denote it \(\mathbf{1}\) — not the number one, but the concept of unity as a closed, bounded, demarcated thing. This is the wall that makes a room possible, the surface that makes a volume possible, the node that makes a standing wave possible. Without demarcation there can be no thing because there is no means for us to distinguish any thing.
In the physical vocabulary of electromagnetism, the structural primitive is \(\varepsilon_0\), the vacuum permittivity. The word permittivity names it precisely: the capacity of the vacuum to permit, contain, and store electric field energy in bounded, enclosed form. The electric field terminates on charges and requires a surface to end on. It lives in closed, node-sustaining configurations. First identified experimentally by Faraday in 1837 as the specific inductive capacity of a medium (Faraday 1838) and given its mathematical form by Maxwell in 1865 (Maxwell 1865), \(\varepsilon_0\) is the constant that governs these configurations — the constant without which no boundary is possible. Therefore \(\varepsilon_0\) is the structural primitive that we herein refer to as \(\mathbf{1}\). This is not an analogy. It is a recognition that the physical definition of \(\varepsilon_0\) and the mathematical definition of \(\mathbf{1}\) are the same statement in two vocabularies.
The second condition is the absence of demarcation: the property that makes passage possible, that admits everything, that has no inside and no outside, no surface and no boundary. We call this the non-structural primitive, and we denote it \(\mathbf{0}\) — not the number zero, but the concept of openness as a complete absence of boundary. The space that makes passage possible. The openness that makes coupling across distance possible. The unbounded that makes the bounded meaningful by contrast.
In the physical vocabulary of electromagnetism, the non-structural primitive is \(\mu_0\), the vacuum permeability. The word permeability names it precisely: the capacity of the vacuum to be permeated — to admit passage, to allow penetration, to offer no surface on which a field can terminate. Magnetic flux has no sources and no sinks. It threads through everything and ends nowhere. Maxwell’s statement \(\nabla \cdot \mathbf{B} = 0\) is not a law imposed on the magnetic field; it is the expression of what \(\mu_0\) is: the open, the unbounded, the structurally incapable of closure. 1 Thus \(\mu_0\) is the non-structural primitive \(\mathbf{0}\). Again, this is not an analogy. The physical definition of \(\mu_0\) and the mathematical definition of \(\mathbf{0}\) are the same statement in two vocabularies.
The structural and non-structural are not opposites in the sense of good and bad, positive and negative, presence and absence. They are poles of a relationship. Neither makes sense without the other. You cannot conceive of a boundary without something on both sides of it. You cannot conceive of unboundedness without something that is bounded to be unbounded relative to. Neither came first. Neither is more fundamental. They are the irreducible pair2.
A careful reader might ask: could we not assign \(\varepsilon_0\) to the non-structural primitive and \(\mu_0\) to the structural? The answer is no, and the reason is not about convention or convenience. It is about the mode of each constant’s action.
\(\varepsilon_0\) governs closed configurations. The energy it stores, \(u_E = \tfrac{1}{2}\varepsilon_0 E^2\), is energy held in a bounded region by a field that terminates on charges. A capacitor — the physical embodiment of \(\varepsilon_0\) — is closed, bounded, and demarcated. \(\varepsilon_0\) is the capacitive constant: it stores by containing.
\(\mu_0\) governs open configurations. The energy it stores, \(u_B = \tfrac{1}{2}\mu_0 H^2\), is energy carried in magnetic flux that loops through space without terminating. An inductor — the physical embodiment of \(\mu_0\) — is open, looping, and boundary-free. \(\mu_0\) is the inductive constant: it stores by threading through.
Containing and threading-through are not the same thing. The identification is forced by what these constants are, not by how we write equations about them. No rewriting of Maxwell’s equations changes the fact that \(\varepsilon_0\) governs closed configurations and \(\mu_0\) governs open ones.
Given the structural primitive \(\mathbf{1}\) and the non-structural primitive \(\mathbf{0}\), one might suppose that the electromagnetic vacuum is now fully specified. It is not. Two inert poles, standing in no relationship, produce nothing. They co-exist in eternal stasis. The structural sits in its closure and the non-structural sits in its openness and nothing passes between them. There is no field, no propagation, and no light.
For anything to change, a third element is required: the relationship that the encounter of the two primitives generates. This relationship is not derived from either primitive alone. It is what comes into existence when the structural meets the non-structural — when \(\mathbf{1}\) is placed in ratio with \(\mathbf{0}\).
Write the encounter as the operation \(\mathbf{1}/\mathbf{0}\). This is the origin of the reciprocal singularity. But before we ask what the result is, we must ask: what kind of thing can the result be?
Here is the observation that determines the entire subsequent structure.
The encounter \(\mathbf{1}/\mathbf{0}\) is not symmetric. To see why, it helps to speak the operation out loud: one is divided by zero. Division requires that the dividend have an interior — something that can be entered, apportioned, opened. A perfectly sealed container cannot be divided into. For \(\mathbf{1}\) to be divided by \(\mathbf{0}\), for the non-structural to enter the structural, there must be something in \(\mathbf{1}\) that permits entry: a crack, a defect, an irremovable asymmetry. A perfectly closed \(\mathbf{1}\) would simply sit in its containment, and \(\mathbf{0}\) would simply sit in its openness, and the operation \(\mathbf{1}/\mathbf{0}\) would be undefined rather than infinite. The break in symmetry — the defect in \(\mathbf{1}\) — is what gives \(1/0 = \infty\) its meaning rather than its impossibility.
One of the two primitives will host this defect — the irremovable asymmetry that makes the process directional rather than stasis-producing. Which one?
The defect cannot live in \(\mathbf{0}\).
A defect is a local departure from an ideal configuration. It is something that is different here than it is everywhere else. But here requires a location. A location requires a boundary — an inside where here is, distinguishable from an outside where it is not. The non-structural primitive \(\mathbf{0}\) has no boundary, no inside, no location. It is uniformly open everywhere. There is no here in \(\mathbf{0}\) where a defect could live.
More precisely: could \(\mathbf{0}\) have more than nothing? More than nothing requires a surplus, and a surplus requires a somewhere in which the surplus resides. But \(\mathbf{0}\) has no somewhere. Then could \(\mathbf{0}\) have less than nothing? No, for less than nothing would require \(\mathbf{0}\) to have a negative structure — which would mean \(\mathbf{0}\) has structure, which contradicts its definition as the non-structural. \(\mathbf{0}\) can only be \(\mathbf{0}\). It admits no defect. It cannot be more or less than what it is.
The defect must live in \(\mathbf{1}\).
The structural primitive \(\mathbf{1}\) has boundaries, surfaces, and locations. A defect in \(\mathbf{1}\) is meaningful and possible: it is a place where the containment is imperfect, where the boundary does not fully close, where the structural has a crack. A cracked container is still a container — it still has an inside and an outside, still demarcates — but it leaks. And that leak is the mechanism by which the process is non-terminal, by which the contained generates the propagating, by which the atom eventually generates the photon. That \(\mathbf{1}\) carries this defect is the sole reason the encounter does not resolve into stasis — the sole reason the universe unfolds in time rather than simply existing as an unchanging geometric fact.
The asymmetry of the encounter \(\mathbf{1}/\mathbf{0}\) is therefore not a choice or a contingency. It is a logical necessity. The defect lives in \(\mathbf{1}\) because only \(\mathbf{1}\) has the interior structure to host it. This is the fact that orients the entire Riemann sphere: the source pole (the site of the defect) is the structural primitive; the sink pole (the site of perfect openness) is the non-structural primitive. Their roles are fixed before any equation is written.
The encounter of the structural with the non-structural is not between equals in the sense of being symmetric. It is between equals in the sense of being co-necessary. Neither is prior or superior. But when they meet in the ratio \(\mathbf{1}/\mathbf{0}\), the result is directional, not symmetric, because only one of them can carry the mark of the encounter. The structural carries it. The non-structural cannot (Theorem 34 in (Coherence Research Collaboration 2026)). This single fact — that \(\mathbf{0}\) cannot host a defect — is what makes the process go in a direction, what breaks the rotational symmetry of the singularity, what initiates propagation rather than stasis, and what gives the residue \(\operatorname{Res}(1/z,\, 0) = 1 \neq 0\) its physical meaning: exactly one unit of structural character persists at the singularity, irremovable, because it is carried by the structural primitive and the structural primitive cannot be made structureless without reducing the entire system to the non-structural alone, which is stasis. The rate at which this process unfolds is \(c\) — constant everywhere it has ever been measured, for reasons that will become clear in Section 1.7.
We now ask: what is the minimal mathematical structure in which the encounter \(\mathbf{1}/\mathbf{0}\) can be represented as a non-terminal directed process? We derive it by successive necessity, attempting each simplification and showing why it fails.
To derive the minimal structure, we must express the encounter \(\mathbf{1}/\mathbf{0}\) as a mathematical map. The natural representation is the reciprocal map \(f(z) = 1/z\), where \(z\) is a complex variable encoding the state of the field at any point in the process: its magnitude \(|z|\) measures the degree of structural character present, and its argument \(\theta = \arg(z)\) encodes the orientation of the process. The question of what minimal space \(z\) can live in is then the question of what minimal mathematical structure the encounter requires.
The simplest possible domain is the real line \(\mathbb{R}\), where \(z\) is just a real number. But on the real line, \(1/z\) has no orientation: approaching zero from the positive side gives \(+\infty\) and from the negative side gives \(-\infty\). The process has no preferred direction. Without direction, there is no propagation — there is only symmetric divergence in two opposing senses, which is a form of stasis.
The minimal domain that supports a directed process with orientation is the complex plane \(\mathbb{C}\): two-dimensional, supporting the angular coordinate \(\theta\) in \(z = re^{i\theta}\). On \(\mathbb{C}\), the reciprocal map gives \(f(z) = (1/r)e^{-i\theta}\): the magnitude diverges and the orientation is reversed. The reversal \(\theta \to -\theta\) is the direction. Two dimensions are necessary.
Left: A perfectly symmetric singularity is electromagnetically inert: \(\mathbf{S} = \mathbf{E}\times\mathbf{H} = \mathbf{0}\) and nothing propagates. Right: A defect (argument reversal \(\arg(1/z)=-\theta\), \(\operatorname{Res}=1\neq 0\)) breaks the symmetry; the vacuum responds via \(Z_0\) so that \(\mathbf{S}\neq\mathbf{0}\) and the directed process \(p\to p^{*}\) begins, developing into the full field structure as depicted in Figure 2.
On the complex plane \(\mathbb{C}\) alone, \(f(z) = 1/z\) as \(z \to 0\) is undefined. The process encounters the singularity and terminates. This is the terminal case: geometry divided by nothingness produces an undefined result, and the process stops.
The non-terminal requirement forces us to adjoin a point at infinity: we extend \(\mathbb{C}\) to \(\hat{\mathbb{C}} = \mathbb{C} \cup \{\infty\}\) by setting \(f(0) = \infty\) and \(f(\infty) = 0\). Thus the singularity is not a wall but a portal, and the process continues through it. This completed space, equipped with the topology of the two-sphere \(S^2\), is the Riemann sphere (Riemann 1851).
The Riemann sphere is not a choice of mathematical setting; it is the unique minimal space in which a non-terminal reciprocal encounter can be continuously represented. Any smaller space terminates the process. Any larger space introduces structure not required by the encounter. The sphere is exact.
The Poincaré–Hopf theorem (Poincaré 1885; Hopf 1926) states: every smooth non-zero vector field on a compact oriented surface without boundary has singular points whose indices sum to the Euler characteristic of the surface. For \(S^2\), the Euler characteristic is \(\chi(S^2) = 2\). The full proof that this forces exactly two poles of index \(+1\) in the minimum realization is given in (Coherence Research Collaboration 2026).
The non-terminal directed process is a non-zero vector field on \(S^2\). It cannot be zero-free — a zero-free field would have index sum zero, contradicting \(\chi(S^2) = 2\). The minimal configuration consistent with the index sum \(= 2\) is exactly two singular points of index \(+1\) each: one source and one sink. Any other configuration either requires a degenerate non-generic tangency (unstable under perturbation) or requires more than two poles (not minimal).
The source pole \(p\) is the site of the defect in \(\mathbf{1}\): the cracked structural primitive, carrying residue \(\operatorname{Res}(1/z,\, 0) = 1\), the measure of the crack. The sink pole \(p^*\) is the antipodal point: the site of perfect openness in \(\mathbf{0}\), receiving without hosting, open without bounding. Their positions are fixed by the topology: \(p\) and \(p^*\) are antipodal on \(S^2\), forced by the spherical symmetry of the underlying space.
Given a point source at \(p\), the potential field between the poles is not a choice. It is determined by the Green’s function of the Laplacian on the sphere. In two dimensions, for a unit point source at the origin, the unique rotationally symmetric solution is: \[G(r) = -\frac{1}{2\pi} \log r, \qquad \phi(r) = \log\frac{1}{r}. \label{eq:logarithmic_potential}\] The logarithmic form is not assumed. It is the only rotationally symmetric solution to the Laplace equation with a point source. The equipotential surfaces of this field are circles at radii \(r\), \(r e^{-\Delta}\), \(r e^{-2\Delta}\), \(\ldots\) for uniform depth steps \(\Delta\) — a geometric sequence, self-similar by construction.
Every simplification either terminates the process, loses the direction, violates the topology, or introduces instability. The Riemann sphere with two antipodal poles and a logarithmic field is the unique minimal structure consistent with a non-terminal, directed, oriented, stable encounter between two irreducible primitives. It is not a model. It is what the encounter necessarily generates.
Genesis of the Riemann sphere from a primitive reciprocal singularity. (i) The argument reversal \(\arg(1/z)=-\theta\) and \(\operatorname{Res}(1/z,0)=1\neq 0\) break the perfect symmetry; the Poynting vector \(\mathbf{S}=\mathbf{E}\times\mathbf{H}\neq\mathbf{0}\) and the directed process begins. (ii) Field lines (geodesics) propagate from \(p\); equipotential circles accumulate near the source; the antipodal sink \(p^{*}\) is forced to appear by the Poincaré–Hopf theorem (Poincaré 1885; Hopf 1926) (\(\chi(S^2)=2\); proved for the present setting in (Coherence Research Collaboration 2026)), shown here as a hollow circle to indicate it is topologically required but not yet fully established. (iii) The complete Riemann sphere: source pole \(p = \infty\) (structural primitive \(\varepsilon_0\), carrying the defect) and sink pole \(p^* = 0\) (non-structural primitive \(\mu_0\)), with full meridional field lines pole-to-pole and logarithmically spaced equipotentials. This is the unique minimal structure consistent with a non-terminal, directed, oriented encounter between the two irreducible primitives.
The logarithmic field between the poles has a natural winding structure. Field lines wind around the source pole \(p\) with some ratio \(\xi\) governing the advance per circuit. The question of what \(\xi\) can be is the question that determines the fine-structure constant, \(\alpha\).
Suppose \(\xi = p/q\) for positive integers \(p\) and \(q\). Then after exactly \(q\) complete circuits around \(p\), the field returns to its precise starting configuration. The process closes. The structural primitive, having generated a finite cycle, returns the geometry to the void. The form that emerged from the encounter dissolves back into the non-structural. In the language of the two primitives: the structural and the non-structural have exchanged their roles a finite number of times and arrived back at the beginning. The distinction between them has collapsed. \(\mathbf{1} = \mathbf{0}\). Nothing persists.
The existence of matter — the empirical fact that stable configurations exist, that atoms hold together, that you are reading this sentence — is the observation that the process has not closed. The winding ratio cannot be rational. \(\xi\) must be irrational, or nothing persists.
Irrationality is necessary but not sufficient. The self-similar logarithmic field has a consistency condition: the ratio \(\xi\) at any depth must equal the ratio at every other depth simultaneously. This requires \(\xi\) to have a purely periodic continued fraction expansion with all partial quotients equal — and the minimum such configuration is all partial quotients equal to \(1\): \[\varphi = [1;\, 1,\, 1,\, 1,\, \ldots] = \frac{1+\sqrt{5}}{2}, \quad \varphi = 1 + \frac{1}{\varphi}. \label{eq:golden_ratio}\] This is the golden ratio. It is the unique irrational that is maximally resistant to rational approximation at every scale simultaneously — a consequence of Hurwitz’s theorem (Hurwitz 1891) and the Markov spectrum theorem (Markov 1879). Every other irrational comes dangerously close to some rational at some scale, making the process nearly-close at that scale, whereas \(\varphi\) never does. It is the most irrational number: the one for which the self-similar field is furthest from any closure at every level of resolution.
Thus the self-similar logarithmic field on \(S^2 \setminus \{p\}\) converges to \(\varphi\) as its unique stable attractor. This is not an optimisation that the field performs. It is the fixed point to which the consistency recursion \(\xi = 1 + 1/\xi\) necessarily converges from any starting value \(\xi_0 > 0\).
If \(\xi = \varphi\) exactly, the process is perfectly non-terminal. The field expands forever, generating infinite form, never producing a closed configuration, and never allowing a bound state to form. There would be propagation without structure — the pure \(Z_0\) field, continuous and undifferentiated, with no bound configurations and no discrete quanta.
For the process to generate stable local closures — the bound configurations we call atoms — the winding must depart slightly from \(\varphi\). There must be some residual tendency toward closure: some small rational approximation that the field nearly satisfies at some scale, around which an almost-closed mode can form. The same defect that initiates the process — the crack in \(\mathbf{1}\), the residue that breaks rotational symmetry — is precisely what displaces the winding ratio below \(\varphi\). One flaw serves both purposes: it makes the universe dynamic, and it makes the universe habitable.
The departure from \(\varphi\) is supplied by the defect in the structural primitive. We established in Section 1.3 that the defect lives in \(\mathbf{1}\) — the structural primitive, the source pole \(p\) — with residue \(\operatorname{Res}(1/z,\, 0) = 1 \neq 0\). This residue is the precise measure of the crack: one unit of structural character persists at the singularity, irremovable, introducing a holonomy (a phase deficit) into every circuit around \(p\).
The holonomy is negative. The orientation reversal \(\theta \to -\theta\) of the reciprocal map reduces the effective winding ratio below the ideal \(\varphi\) such that the actual winding ratio is: \[\xi_{\text{actual}} = \varphi - \alpha, \label{eq:actual_winding}\] where \(\alpha\) is assembled from four facts established in the companion proof paper (Coherence Research Collaboration 2026):
Assembling: \[\alpha = \frac{Z_0}{2 R_K} = \frac{\sqrt{\mu_0/\varepsilon_0}}{2h/e^2} = \frac{e^2}{2\varepsilon_0 h c} \approx \frac{1}{137.036}. \label{eq:alpha}\] (Theorem 37 in (Coherence Research Collaboration 2026).)
\(\alpha\) is the fine-structure constant. It is not a coupling constant whose value was measured and thereafter accepted as a brute fact of nature. It is the numerical signature of the crack in the structural primitive: the specific measure of how far the relationship between the structural and non-structural departs from perfect golden-ratio non-closure. Its value is determined by what \(\mathbf{1}\) and \(\mathbf{0}\) are, by the topology of the sphere they generate, and by the minimum quantum of the closed process. No free parameter enters.
\(\alpha \approx 1/137\) means the departure from \(\varphi\) is small. This is necessary on both ends:
If \(\alpha\) were zero, the winding ratio would be exactly \(\varphi\) and no atoms would form. The process would be perfectly non-terminal, generating propagation without bound configurations.
If \(\alpha\) were large — comparable to \(1\) — the departure from \(\varphi\) would be large, the process would be nearly rational at many scales, and bound configurations would be so short-lived as to be unrecognizable as stable matter.
The specific value \(\approx 1/137\) places the vacuum in the regime where: the process is irrational enough to sustain matter indefinitely (atoms do not spontaneously dissolve), and rational enough to allow change (atoms can radiate and transition). The fine-structure constant is the measure of the narrow window in which stable, radiating matter is possible.
We pause here to address a deep question that the preceding development raises: if the electromagnetic vacuum is generated by the encounter of two primitives, what is it generated in? What is the prior space that contains the Riemann sphere?
The question is malformed, and understanding why it is malformed is essential.
The standard picture of physics treats the electromagnetic vacuum as a background — a fixed stage with measured properties (\(\varepsilon_0\), \(\mu_0\), \(c\), \(Z_0\)) against which particles, fields, and interactions perform. In this picture, the vacuum is prior to the physics: it is already there, already equipped with its constants, and physics happens inside it.
This paper takes a different position, and it is not merely a philosophical preference. It is forced by the empirical constancy of \(c\). The speed of light has been measured everywhere accessible to human observation: in laboratory vacuum today, in the spectra of distant galaxies billions of light-years away, in the cosmic microwave background encoding the state of the universe 380,000 years after its origin. In every measurement, at every scale, at every epoch: \(c\) is the same.
Not approximately the same. The same.
If the electromagnetic vacuum were a stage — a pre-existing container whose properties were set at some moment and thereafter maintained — then in principle those properties could vary. A maintained property can change if the maintaining mechanism is disturbed. Physical stages can be redecorated. What never varies under any physical process is not a property of a stage but a consequence of a relationship between primitives that are prior to any physical process.
\(c\) is constant because \(\varepsilon_0\) and \(\mu_0\) are constant. \(\varepsilon_0\) and \(\mu_0\) are constant because they are not physical quantities that can vary: they are the structural and non-structural primitives whose encounter generates physical quantities. The structural cannot become more or less structural. The non-structural cannot become more or less open. Their ratio \(Z_0\) is therefore fixed. And therefore \(c = 1/\sqrt{\varepsilon_0\mu_0}\) is fixed — not as a postulate of special relativity, but as a consequence of the primitives being what they are.
The constancy of \(c\) is the empirical proof that the electromagnetic vacuum is not a stage. It is the process itself, continuously unfolding at the rate set by the relationship between the two primitives, without interruption, from the original encounter to the present measurement. Every photon propagating right now is the relationship between \(\mathbf{1}\) and \(\mathbf{0}\), propagating. Every atom holding together right now is a local closure within that propagating relationship. The vacuum is not the container of physics. It is the physics, continuously generating itself.
We have derived, from two irreducible primitives and the logical necessity of their encounter, the following chain:
Everything that follows in this paper is the consequence of reading this eight-step chain with the instruments of atomic physics. The next section shows that the orbital impedance \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) follows from applying Coulomb’s law and the Biot–Savart law to the electromagnetic closure mode at quantum number \(n\). The section after that introduces the \(\kappa\) coordinate: the natural depth ruler for atomic spectra, in which the logarithmic hierarchy of the Riemann sphere field is directly readable. The remainder of the paper traces the chain from the original encounter to the spectral record of every atom in the periodic table — the gap at \(IE/4\), confirmed across 71 ions with zero free parameters and zero falsifications.
Section 1 derived the electromagnetic vacuum from two irreducible primitives and the logical necessity of their encounter. We showed that the three-element composition — structural primitive, non-structural primitive, and the relationship their encounter generates — is the minimum consistent with a non-terminal, directed, oriented process. The Riemann sphere, the logarithmic field, the golden ratio, and the fine-structure constant \(\alpha\) all followed from that minimum structure by necessity.
Maxwell’s equations now enter. Not as the foundation — Section 1 is the foundation — but as the interior description: the precise, elegant, experimentally confirmed account of what happens inside of the container that Section 1 derived. Maxwell’s equations describe the field that the encounter of the two primitives generates, using the vocabulary of curl relations and constitutive constants, and demonstrate perfect agreement with the primitive structure. Maxwell’s equations could not be otherwise, because they are describing the interior of the same three-element structure — the same structural primitive, non-structural primitive, and relationship between them — seen now from within rather than derived from without.
We are now inside our universe.
This section has two tasks. The first is to show that Maxwell’s equations encode the three-element structure exactly, and that both vacuum constants are necessary to do so. The second is to identify the specific representational choice that has made the third element — the relationship \(Z_0\) — invisible to most of physics for most of its history. That choice is not an error in Maxwell. It is a compression that is exact for one class of questions and blind to another. Making the compression explicit is the first step toward restoring what it discards.
Maxwell’s equations in vacuum, in the absence of free charges and currents, are: \[\begin{aligned} \nabla \cdot \mathbf{E} &= 0, \label{eq:gauss_E} \\ \nabla \cdot \mathbf{B} &= 0, \label{eq:gauss_B} \\ \nabla \times \mathbf{E} &= -\mu_0 \frac{\partial \mathbf{H}}{\partial t}, \label{eq:faraday} \\ \nabla \times \mathbf{H} &= \varepsilon_0 \frac{\partial \mathbf{E}}{\partial t}. \label{eq:ampere} \end{aligned}\] Four equations. Two constants. The constants enter separately and play structurally distinct roles.
Read these equations in light of what we established in Section 1.
Equation [eq:gauss_E] says the electric field has no sources in the vacuum: \(\nabla \cdot \mathbf{E} = 0\). This is the statement that the structural primitive \(\varepsilon_0\) governs a field that lives in closed, node-sustaining configurations when no charges are present. The electric field terminates on charges precisely because \(\varepsilon_0\) is the containing primitive. In free space, with no charges, the containing closes on itself: no sources, no sinks, divergence zero.
Equation [eq:gauss_B] says the magnetic field has no sources anywhere, ever: \(\nabla \cdot \mathbf{B} = 0\). This is the non-structural primitive speaking directly. The magnetic field threads through without terminating. It has no somewhere to end. It passes through everything because \(\mu_0\) is the open, passage-admitting, non-structural primitive. \(\nabla \cdot \mathbf{B} = 0\) is not a law imposed on the magnetic field from outside. It is the definition of what \(\mu_0\) is.
Equations [eq:faraday] and [eq:ampere] are the mutual pursuit: a changing electric field drives a curl in the magnetic field, and a changing magnetic field drives a curl in the electric field. Neither leads. Each is the other’s cause. The constants \(\mu_0\) and \(\varepsilon_0\) appear as the coefficients of this mutual coupling — \(\mu_0\) governing how strongly the changing electric field drives the magnetic response, \(\varepsilon_0\) governing how strongly the changing magnetic field drives the electric response. These roles are not symmetric and not interchangeable. They are the structural and non-structural primitives, each governing the coupling in the direction appropriate to its nature.
Reading Maxwell’s equations in light of Section 1:
\(\nabla \cdot \mathbf{E} = 0\): The structural primitive (\(\varepsilon_0\), the containing mode) produces a field that closes on itself in free space. No sources, no sinks. Form without escape.
\(\nabla \cdot \mathbf{B} = 0\): The non-structural primitive (\(\mu_0\), the passage-admitting mode) produces a field that never terminates. No sources, no sinks. Passage without boundary.
\(\nabla \times \mathbf{E} = -\mu_0 \partial_t \mathbf{H}\): The changing structural field (E) drives the non-structural response (H) at a rate governed by \(\mu_0\), the openness to penetration.
\(\nabla \times \mathbf{H} = \varepsilon_0 \partial_t \mathbf{E}\): The changing non-structural field (H) drives the structural response (E) at a rate governed by \(\varepsilon_0\), the capacity for containment.
These are not four separate laws. They are one process — the mutual pursuit of the structural and non-structural primitives — written in the vocabulary of curl equations.
From equations [eq:faraday] and [eq:ampere], taking the curl of each and substituting, one derives the wave equations: \[\nabla^2 \mathbf{E} = \mu_0 \varepsilon_0 \frac{\partial^2 \mathbf{E}}{\partial t^2}, \qquad \nabla^2 \mathbf{H} = \mu_0 \varepsilon_0 \frac{\partial^2 \mathbf{H}}{\partial t^2}. \label{eq:wave_equations}\] The coefficient \(\mu_0 \varepsilon_0\) appears as a product. Maxwell identified this product with the propagation speed: \[c^2 = \frac{1}{\mu_0 \varepsilon_0}. \label{eq:c_squared}\] This step is exact for propagation. The wave equations depend only on the product, and the product is \(1/c^2\). But it is many-to-one: for any fixed \(c\), infinitely many distinct pairs \((\varepsilon_0, \mu_0)\) satisfy \(\varepsilon_0 \mu_0 = 1/c^2\). Each such pair shares the same propagation speed and differs in one quantity: their ratio.
The ratio is the vacuum impedance: \[Z_0 = \sqrt{\frac{\mu_0}{\varepsilon_0}}. \label{eq:Z0_definition}\] \(Z_0\) is not derived from \(c\). It is independent of \(c\). Together, \(c\) and \(Z_0\) recover both constants uniquely: \[\mu_0 = \frac{Z_0}{c}, \qquad \varepsilon_0 = \frac{1}{Z_0\, c}. \label{eq:recover_constants}\] Neither \(c\) alone nor \(Z_0\) alone is sufficient. Their pair is exact and unique. Maxwell’s equations contain two independent invariants. The passage to the wave equation retains one and suppresses the other.
This suppressed invariant is the third element of Section 1: the relationship \(Z_0 = \sqrt{\mu_0/\varepsilon_0}\) is the ratio of the non-structural primitive to the structural primitive, the field between the poles of the Riemann sphere, propagating. Maxwell’s equations have always contained it. The passage to \(c^2\) discards it. This paper is, among other things, the restoration of what that passage discards.
The passage \((\varepsilon_0, \mu_0) \mapsto c^2\) is not an error. It is a representational choice, and like all representational choices it is exact for some questions and silent on others.
For propagation questions — how fast does the wave travel, what is the dispersion relation, what are the wave equations — the compression is exact. Every quantity that depends only on the product \(\varepsilon_0 \mu_0\) is fully determined by \(c\). The richness of wave optics, electrodynamics, and classical field theory lives in this territory, and \(c\) is the right tool for all of it.
For partition questions — how is field energy divided between the electric mode and the magnetic mode, what is the amplitude ratio \(|\mathbf{E}|/|\mathbf{H}|\) at a given point, what impedance does a given field configuration present to the vacuum — the compression is silent. These quantities depend on the ratio \(\mu_0/\varepsilon_0 = Z_0^2\), which is suppressed by the passage to \(c^2\). No amount of propagation information recovers it.
The cost of the compression becomes visible in a specific class of physical situations: those where the field configuration at a boundary matters, not just the propagation speed. Impedance matching at an interface. Power transfer between a source and a load. The partition of field energy between capacitive and inductive modes in a resonant structure. And — as this paper establishes — the partition of transition energies in atomic spectra between the capacitive sector below \(IE/4\) and the inductive sector above it.
All of these are partition questions. All of them are invisible in a \(c\)-only description. All of them become visible when \(Z_0\) is restored as an independent invariant.
Maxwell’s four equations as interior description of the Riemann sphere. The sphere is the same structure derived in Section 1: north pole \(p = \varepsilon_0\) (structural primitive \(\mathbf{1}\), source of the directed process) and south pole \(p^{*} = \mu_0\) (non-structural primitive \(\mathbf{0}\), sink). The red great circle is the \(\mathbf{E}\)-mode: the electric field closes on charges (\(\nabla\cdot\mathbf{E}=0\) in free space) and its curl drives the \(\mathbf{H}\)-mode (\(\nabla\times\mathbf{E}=-\mu_0\,\partial_t\mathbf{H}\)). The blue great circle is the \(\mathbf{H}\)-mode: the magnetic field threads through without terminating (\(\nabla\cdot\mathbf{B}=0\) always) and its curl drives the \(\mathbf{E}\)-mode (\(\nabla\times\mathbf{H}=\varepsilon_0\,\partial_t\mathbf{E}\)). Orthogonality of \(\mathbf{E}\) and \(\mathbf{H}\) is not imposed; it is the topological consequence of two great circles on a sphere. Their intersection (open circles) is the unique configuration where \(|\mathbf{E}|/|\mathbf{H}|=Z_0\): the impedance-matched free-photon mode, the third element propagating at \(c\). Maxwell’s equations do not introduce new structure; they describe the same three-element geometry from within (Theorem 34 in (Coherence Research Collaboration 2026)).
The simplest and most direct expression of \(Z_0\) as a physical invariant is in the free-propagating electromagnetic mode: the photon.
For a plane wave propagating in the \(\hat{k}\) direction in vacuum, the plane-wave solutions to equations [eq:faraday] and [eq:ampere] require: \[\mathbf{H} = \frac{\hat{k} \times \mathbf{E}}{Z_0}, \qquad \text{so} \qquad \frac{|\mathbf{E}|}{|\mathbf{H}|} = Z_0 \label{eq:photon_impedance}\] at every point, at every moment, for every frequency, for every polarization. The ratio of the electric to magnetic field amplitude in a free photon is exactly \(Z_0\). Always. Everywhere. Without exception.
This is the physical meaning of \(Z_0\) stated as concisely as possible. A propagating photon is the pure expression of the relationship between the structural and non-structural primitives — the third element — carrying the ratio of the non-structural to the structural at every point. The photon does not travel through the vacuum. It is the vacuum, propagating: the relationship between \(\mathbf{1}\) and \(\mathbf{0}\), moving at \(c\), maintaining the ratio \(Z_0\) between the electric and magnetic amplitudes as it goes.
Any departure from \(Z_0\) in a field configuration is a departure from the free-propagating state. Bound fields depart from \(Z_0\). They must: a bound field is a field that has not propagated away. Its impedance is different from \(Z_0\), and the degree of mismatch determines how hard it is for the field to transition from the bound state to the free-propagating state — that is, how hard it is for the atom to emit a photon, and what energy that photon carries. The energy of the emitted photon is \(\Delta E = h\nu\), where \(h\) is not a postulate but the scaling invariant of the self-similar logarithmic field derived in Section 3.10.
This is the content of Section 3, which derives the orbital impedance \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) from Coulomb’s law and the Biot–Savart law. The mismatch \(Z_{\rm orb}/Z_0 = n/\alpha \approx 137n\) is the same \(\alpha\) that Section 1 identified as the departure from the golden-ratio attractor. The spectrometer reading it as a fine-structure correction and the circuit model reading it as an impedance ratio are two instruments measuring the same quantity.
The compression \((\varepsilon_0, \mu_0) \mapsto c^2\) has a seductive feature that makes it easy to overlook what it discards. It appears to factor \(c\) into two constituent pieces corresponding to the two primitives: \[c^2 = \frac{1}{\varepsilon_0 \mu_0} = \underbrace{\frac{1}{\varepsilon_0}}_{c_E^2} \cdot \underbrace{\frac{1}{\mu_0}}_{c_M^2}, \label{eq:factoring_attempt}\] where \(c_E = 1/\sqrt{\varepsilon_0}\) and \(c_M = 1/\sqrt{\mu_0}\) look like individual speeds for the electric and magnetic contributions.
They are not. Their units are incompatible: \[= \mathrm{m\,s^{-1}\,C^{-1}\,kg^{1/2}}, \qquad [c_M] = \mathrm{m\,s^{-1}\,C\,kg^{-1/2}}. \label{eq:unit_mismatch}\] Only the product \(c_E \cdot c_M\) recovers the dimensions of a speed. The factoring is not unique: any rescaling \(c_E \to \lambda c_E\), \(c_M \to c_M/\lambda\) leaves \(c\) unchanged while altering \(\varepsilon_0\) and \(\mu_0\) individually. Fixing \(c\) fixes the product \(\varepsilon_0\mu_0\) and leaves the ratio \(\mu_0/\varepsilon_0\) — which is \(Z_0^2\) — completely free.
This is the fiber structure of the compression. The passage to \(c^2\) is a projection from the two-dimensional space \((\varepsilon_0, \mu_0)\) onto a one-dimensional line. Each point on that line corresponds to an infinite family of distinct vacua, all sharing the same propagation speed and all differing in \(Z_0\). The factoring trap is the illusion that this projection is invertible. It is not.
The curl equations show precisely what the compression discards. In Maxwell’s equations [eq:faraday] and [eq:ampere], the two primitives appear separately and play asymmetric roles: \(\mu_0\) governs how a changing electric field drives the magnetic response, and \(\varepsilon_0\) governs how a changing magnetic field drives the electric response. Their individual appearances in the curl equations are not redundant — they fix the ratio \(|\mathbf{E}|/|\mathbf{H}| = Z_0\) in every admissible wave solution (Section 2.4). When the wave equation is derived by combining the two curl equations, the product \(\varepsilon_0\mu_0 = 1/c^2\) survives and the ratio \(\mu_0/\varepsilon_0 = Z_0^2\) disappears. The compression is not an accident of notation: it is built into the act of combining the two curl equations into one wave equation. What is lost is the separate structural role of each primitive — and with it, the partition geometry that determines how field energy divides between the electric and magnetic modes at every boundary, every interface, and every orbital closure.
The decomposition becomes unique only when the second independent invariant is supplied. Given both \(c\) and \(Z_0\), equations [eq:recover_constants] recover \(\varepsilon_0\) and \(\mu_0\) uniquely: \[\mu_0 = \frac{Z_0}{c}, \qquad \varepsilon_0 = \frac{1}{Z_0\,c}. \label{eq:recover_constants_repeated}\] The pair \((c,\, Z_0)\) is therefore the complete and minimal description of the electromagnetic vacuum: \(c\) fixes the product \(\varepsilon_0\mu_0\) (propagation speed, dispersion), and \(Z_0\) fixes the ratio \(\mu_0/\varepsilon_0\) (field-energy partition, impedance). Neither alone is sufficient; together they are exact.
The passage from the full pair \((\varepsilon_0,\,\mu_0)\) to the single scalar \(c^2\) is therefore a many-to-one compression: it retains the propagation invariant and discards the partition invariant \(Z_0\) — the third element of the primitive structure identified in Section 1. The compression is exact for any question that depends only on how fast the field propagates, and silent on any question that depends on how field energy is divided between the electric and magnetic modes. The proof that this suppression is structural rather than merely notational, and that restoring \(Z_0\) resolves the underdetermination without modifying Maxwell’s equations, is given in (K. B. Heaton and Coherence Research Collaboration 2026b). That incompleteness — and its consequences for atomic spectra — is the subject of this paper.
We can now state the complete correspondence between the primitive structure of Section 1 and Maxwell’s equations, in a form that will serve as the reference point for every subsequent section.
The electromagnetic vacuum is generated by two irreducible primitives and their relationship. Maxwell’s equations describe the interior of that generation. The correspondence is exact:
|
Primitive structure |
Physical constant |
Role in Maxwell |
|
Structural primitive \(\mathbf{1}\) |
\(\varepsilon_0\) |
Governs containing mode; \(\nabla\cdot\mathbf{E}=0\) in free space |
|
Non-structural primitive \(\mathbf{0}\) |
\(\mu_0\) |
Governs passage mode; \(\nabla\cdot\mathbf{B}=0\) always |
|
Relationship (third element) |
\(Z_0 = \sqrt{\mu_0/\varepsilon_0}\) |
\(|\mathbf{E}|/|\mathbf{H}| = Z_0\) in every free photon |
|
Process rate |
\(c = 1/\sqrt{\mu_0\varepsilon_0}\) |
Wave propagation speed; the product invariant |
|
Crack in \(\mathbf{1}\) |
\(\alpha = Z_0/2R_K\) |
Departure from \(\varphi\); governs impedance mismatch |
The first three rows are the three elements. The fourth is the rate at which the process unfolds — constant, because the primitives are constant. The fifth is the measure of the crack in the structural primitive — the quantity whose consequences for atomic spectra occupy the remainder of this paper.
Maxwell’s equations are not the foundation of this paper, but the interior description of the foundation established in Section 1. They describe, in the vocabulary of curl relations and constitutive constants, the field that the encounter of the two primitives necessarily generates.
The passage from Maxwell’s full two-constant description to the one-constant wave equation is a compression that is exact for propagation and silent on partition. The suppressed coordinate is \(Z_0\) — the third element, the relationship between the structural and non-structural primitives, the ratio that every free photon carries at every point.
Restoring \(Z_0\) as an independent invariant does not modify Maxwell’s equations. It reads them more completely. The next section applies that complete reading to the bound electron — the field that has not propagated away — and derives the orbital impedance from which everything else follows.
Section 2 reveals that Maxwell’s equations contain two independent invariants, that passing from \((\varepsilon_0, \mu_0)\) to \(c^2\) suppresses \(Z_0\), and that a free photon carries the ratio \(|\mathbf{E}|/|\mathbf{H}| = Z_0\) at every point. The photon is the matched mode of the vacuum. The question we now ask is: what is happening to the structure that has not propagated away? What is the electromagnetic character of the field that is bound inside an atom?
We now derive the answer from first principles, using only Coulomb’s law and the Biot–Savart law. No quantum mechanical postulates beyond the closure velocity \(v_n = \alpha c/n\) are needed. No free parameters enter. The result — the orbital field impedance \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) — follows by necessity from the field equations applied to the closure condition.
We then show that this result places the orbital boundary into an exact Thévenin circuit, with the nucleus as the electromotive source, the orbital mode as the internal impedance, and the vacuum as the load. Thus the circuit is not a model; it is the complete electromagnetic description of the orbital–vacuum interface.
The standard picture of the atom — a nucleus at the center, electrons occupying probability distributions described by wavefunctions — is correct and we do not ask the reader to abandon it. But for the purpose of this section, we need to look at the same object from a different angle: not as a particle in a potential, but as an electromagnetic field configuration.
A bound electron at principal quantum number \(n\) is a self-consistent electromagnetic closure mode of the Coulomb field. The word closure is key. The electromagnetic field of the electron, driven by the nuclear Coulomb potential, organizes itself into a configuration that returns to its starting state after one complete cycle. It closes on itself. The closure is what makes it stable: an electromagnetic configuration that does not close simply propagates away.
The closure condition imposes a definite velocity on the field: \[v_n = \frac{\alpha c}{n}, \label{eq:closure_velocity}\] where \(\alpha\) is the fine-structure constant and \(n\) is the principal quantum number. This is not a Bohr model assumption. It is confirmed independently by quantum mechanics: the expectation value of the electron velocity in the exact hydrogen wavefunctions, computed from the virial theorem applied to the Coulomb potential, gives \(\langle v \rangle_n = \alpha c / n\) exactly. The present derivation depends on the velocity, not on any picture of what is carrying it.
The closure also imposes a definite radius: \[r_n = n^2 a_0, \label{eq:orbital_radius}\] where \(a_0 = \hbar/(m_e \alpha c) \approx 52.9\,\mathrm{pm}\) is the Bohr radius. Together, \(v_n\) and \(r_n\) specify the kinematics of the closure mode. They are the two inputs to the derivation that follows.
The wavefunction description of the atom is exact for computing energy eigenvalues, transition frequencies, and selection rules. For those purposes, it is the right tool.
The closure picture is the right tool for a different purpose: computing the field impedance of the bound configuration — the ratio of electric to magnetic field amplitude that the orbital presents to the vacuum. This ratio depends on the dynamics of the field at the orbital boundary, not on the probability distribution of the electron. It is an electromagnetic quantity, and the electromagnetic description — the Coulomb field, the orbital current, the Biot–Savart magnetic field — is the right language for it.
The two descriptions are not in conflict; they are two coordinate systems for the same physical object. The closure picture gives us \(Z_{\rm orb}\); the wavefunction picture gives us the energy levels. Both are needed, and later in this paper we will need both.
The Coulomb potential of the nucleus produces a radial electric field at the orbital radius \(r_n\). By Coulomb’s law: \[|\mathbf{E}_n| = \frac{e}{4\pi\varepsilon_0\, r_n^2}. \label{eq:coulomb_field}\] Every symbol: \(e\) is the elementary charge, \(\varepsilon_0\) is the vacuum permittivity, and \(r_n = n^2 a_0\) is the orbital radius at quantum number \(n\). The field points radially outward (for a positive nucleus). Its magnitude depends on \(r_n\) and on the charge \(e\), but on nothing else.
Substituting \(r_n = n^2 a_0\): \[|\mathbf{E}_n| = \frac{e}{4\pi\varepsilon_0\, n^4 a_0^2}. \label{eq:electric_field_explicit}\] This is the electric field of the bound configuration at the orbital radius. It is the standard Coulomb result. Nothing new here yet.
The closure mode at quantum number \(n\) involves charge circulating at velocity \(v_n\) in a loop of radius \(r_n\). This circulating charge constitutes an electric current, and that current generates a magnetic field.
The charge \(e\) completes one circuit of the loop in time \(t = 2\pi r_n / v_n\). The current is charge per unit time: \[I_n = \frac{e}{t} = \frac{e\, v_n}{2\pi r_n} = \frac{e\, (\alpha c/n)}{2\pi\, n^2 a_0} = \frac{e\,\alpha c}{2\pi\, n^3 a_0}. \label{eq:orbital_current}\] This is the orbital current. It is entirely determined by \(e\), \(\alpha\), \(c\), \(n\), and \(a_0\) — no free parameters.
A circular current loop of radius \(r_n\) carrying current \(I_n\) produces a magnetic field at its center. By the Biot–Savart law, this field is: \[|\mathbf{H}_n| = \frac{I_n}{2\, r_n}. \label{eq:biot_savart}\] Substituting the expressions for \(I_n\) and \(r_n\): \[|\mathbf{H}_n| = \frac{e\,\alpha c}{2\pi\, n^3 a_0} \cdot \frac{1}{2\, n^2 a_0} = \frac{e\,\alpha c}{4\pi\, n^5 a_0^2}. \label{eq:magnetic_field_explicit}\] This is the magnetic field generated by the orbital current at the loop center. Again, entirely determined by known quantities and no free parameters.
We now have the electric field and the magnetic field of the closure mode at quantum number \(n\). We form their ratio. This ratio is the field impedance of the orbital — the electromagnetic impedance that the bound configuration presents to the vacuum at its boundary.
\[\begin{aligned} Z_{\rm orb}(n) &\equiv \frac{|\mathbf{E}_n|}{|\mathbf{H}_n|} \nonumber \\[6pt] &= \frac{e/(4\pi\varepsilon_0\, n^4 a_0^2)} {e\,\alpha c/(4\pi\, n^5 a_0^2)} \nonumber \\[6pt] &= \frac{e}{4\pi\varepsilon_0\, n^4 a_0^2} \cdot \frac{4\pi\, n^5 a_0^2}{e\,\alpha c} \nonumber \\[6pt] &= \frac{n}{\varepsilon_0\,\alpha c}. \label{eq:ratio_intermediate} \end{aligned}\] Watch what happened in the second-to-third line: \(e\) cancels, \(a_0^2\) cancels, and \(4\pi\) cancels. The ratio of the electric field to the magnetic field at the orbital radius is completely independent of both the orbital radius \(r_n\) and the elementary charge \(e\). These two quantities, which dominate the individual fields, vanish from the ratio.
What remains is \(n/(\varepsilon_0\,\alpha c)\). Noting the standard identity \[\frac{1}{\varepsilon_0\, c} = \mu_0 c = \sqrt{\frac{\mu_0}{\varepsilon_0}} = Z_0,\] where the first equality uses \(\mu_0\varepsilon_0 = 1/c^2\), one obtains immediately \[\boxed{Z_{\rm orb}(n) = \frac{Z_0\, n}{\alpha}.} \label{eq:Zorb}\]
This is the orbital field impedance. It follows from Coulomb’s law and the Biot–Savart law, with no free parameters and no model assumptions beyond the closure velocity \(v_n = \alpha c/n\). No quantum mechanics beyond the virial theorem has been used. This is the electromagnetic field ratio of the bound configuration, derived from the field equations.
At \(n = 1\), the ground state of hydrogen: \[Z_{\rm orb}(1) = \frac{Z_0}{\alpha} \approx \frac{376.730\,\Omega}{7.2974 \times 10^{-3}} \approx 51{,}626\,\Omega. \label{eq:Zorb_n1}\] The vacuum has \(Z_0 \approx 377\,\Omega\). The ground state of hydrogen has \(Z_{\rm orb}(1) \approx 51{,}626\,\Omega\). The bound field is approximately \(137\) times more impedance-mismatched to the vacuum than the vacuum is to itself.
At \(n = 2\): \(Z_{\rm orb}(2) \approx 103{,}252\,\Omega\). At every higher quantum number, the mismatch grows. There is no orbital, at any principal quantum number, that is impedance-matched to the vacuum.
Let the reader sit with this for a moment. The electric field of the orbital is far too strong relative to its magnetic field compared to what the vacuum expects. The vacuum expects \(|\mathbf{E}|/|\mathbf{H}| = 377\,\Omega\). The orbital offers \(51{,}626\,\Omega\) at its boundary. That discrepancy — by a factor of precisely \(1/\alpha \approx 137\) at \(n=1\) — is not a curiosity. It is the fundamental electromagnetic reason that bound electrons do not spontaneously radiate. We will make this precise in the next subsection.
The orbital field impedance \(Z_{\rm orb}(n)\) tells us how the bound field differs from the free field. But it does not yet tell us the complete picture of the orbital–vacuum interface. For that, we need to identify the three electromagnetic elements that constitute the interface, and show that they form an exact circuit. The three elements are as follows:
The Coulomb potential of the nucleus is the source of the electromotive force that drives the orbital field. The relevant quantity is the open-circuit voltage: the work required to remove the electron from the orbital to the continuum when no current flows through the orbital load. This is the ionization energy \(IE\) of the ion. In circuit language: the nucleus is an ideal voltage source with open-circuit voltage \(V_{\rm TH} = IE\).
Two points of precision. First, \(V_{\rm TH} = IE\) is the voltage at the terminals when the load is disconnected — when the electron is not oscillating, not radiating, not doing anything except sitting at the boundary. The ionization energy is exactly this: the work required to separate the electron from the nucleus at zero velocity, with nothing connected. Second, the nucleus is the source. The electron is the load — the reactive mode that the source drives.
The closure mode at quantum number \(n\) presents the field impedance \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) to the vacuum. In circuit language, this is the internal impedance of the source — the series impedance between the electromotive force and the output terminals. In the Thévenin representation, \(Z_{\rm orb}\) appears as a series element in the top wire: the field must pass through this impedance before reaching the load. The magnitude of \(Z_{\rm orb}\) relative to the load \(Z_0\) determines how much of the source voltage appears across the load — and how much power can be transferred.
The vacuum is not a passive bystander. It is the transmission medium through which the field propagates when it leaves the orbital, and it is the load that the orbital must drive when it emits a photon. Its impedance is \(Z_0 \approx 377\,\Omega\). In circuit language, \(Z_0\) is the load resistance connected across the output terminals of the Thévenin source. The fraction of source power delivered to this load is determined by how well \(Z_0\) matches the internal impedance \(Z_{\rm orb}\).
These three elements — source \(V_{\rm TH} = IE\), internal impedance \(Z_{\rm orb}\), and load \(Z_0\) — constitute a Thévenin circuit. Figure 4 shows the circuit in its Thévenin and Norton forms.
Thévenin (left) and Norton (right) equivalent circuits of the orbital–vacuum interface. The nucleus is the electromotive source (\(V_\mathrm{TH} = IE\)); the orbital mode is the reactive internal impedance (\(Z_\mathrm{orb} = Z_0\, n/\alpha\)); the vacuum is the transmission medium and load (\(Z_0\)). The two representations are dual: the Thévenin reads the boundary from the E-sector (voltage, capacitive terminus); the Norton reads it from the H-sector (current, inductive onset). The product \(V_\mathrm{TH} \cdot I_N = IE^2/Z_\mathrm{orb}\) is the power available at the boundary, independent of which representation is used.
The Thévenin circuit is exact, not approximate. Thévenin’s theorem states that any two-terminal electromagnetic interface — no matter how complex its internal structure — can be represented as a voltage source in series with an impedance. The orbital–vacuum interface has two terminals (the E-mode rail at the top and the H-mode rail at the bottom), a source (\(V_{\rm TH} = IE\)), and an internal impedance (\(Z_{\rm orb}\)). Thévenin’s theorem applies. Therefore, the circuit is not a model but the complete description.
Two features of the Thévenin circuit deserve attention:
In a standard Thévenin circuit, the internal element is a resistance \(R_S\) that dissipates power. The orbital impedance \(Z_{\rm orb}\) is a reactive impedance: it stores energy rather than dissipating it. This is why the orbital is stable. A dissipative internal element would drain the source; a reactive one stores the energy in the field and returns it to the source. The bound electron’s stability is not a consequence of a selection rule; it is a consequence of its internal impedance being reactive rather than resistive.
The top wire carries the E-mode: the electric field, organized by the structural primitive \(\varepsilon_0\). The bottom wire carries the H-mode: the magnetic field, organized by the non-structural primitive \(\mu_0\). In the diagram, the bottom wire appears as a short circuit. This is not an ontological claim. It is a representational compression: both modes are already encoded in the \(Z_0\) load box, which carries their ratio \(\sqrt{\mu_0/\varepsilon_0}\). The H-mode is real and physical; it is \(\mu_0\); it is the non-structural primitive. It is represented once, in \(Z_0\), rather than twice.
With the Thévenin circuit in hand, the reflection coefficient at the orbital–vacuum interface follows immediately from transmission-line theory.
When a field traveling through a medium of impedance \(Z_{\rm orb}\) encounters a medium of impedance \(Z_0\), the fraction of the field amplitude that is reflected back is: \[\Gamma(n) = \frac{Z_{\rm orb}(n) - Z_0}{Z_{\rm orb}(n) + Z_0}. \label{eq:reflection_coeff}\] Substituting \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) and simplifying: \[\Gamma(n) = \frac{Z_0\, n/\alpha - Z_0}{Z_0\, n/\alpha + Z_0} = \frac{n/\alpha - 1}{n/\alpha + 1} = \frac{n - \alpha}{n + \alpha} \;\longrightarrow\; 1 \quad \text{for all } n \geq 1. \label{eq:reflection_coeff_simplified}\] At \(n = 1\): \(\Gamma(1) = (1 - \alpha)/(1 + \alpha) \approx (1 - 0.0073)/(1 + 0.0073) \approx 0.9855\).
Nearly total reflection. At every principal quantum number, at every orbital, the reflection coefficient approaches \(1\). The orbital–vacuum interface is one of the most highly reflective boundaries in nature.
What does \(\Gamma \to 1\) mean physically? A reflection coefficient of \(+1\) means total reflection with no phase reversal: the reflected wave returns in phase with the incoming wave. The field energy is stored, not radiated. The orbital does not radiate not because of a quantum mechanical selection rule or a postulate about stationary states. It does not radiate because the impedance mismatch between the orbital field and the vacuum is so severe that nearly all the field energy bounces back from the boundary. The orbital is a near-perfect resonant trap.
The power transmission factor is the complement: \[T(n) = 1 - \Gamma^2(n) = \frac{4\alpha\, n}{(n+\alpha)^2} \approx \frac{4\alpha}{n} \quad (\alpha \ll n). \label{eq:transmission}\] At \(n = 1\), \(T(1) \approx 4\alpha \approx 0.029\): about \(2.9\%\) of the field energy can escape the orbital at any given attempt. The factor \(\alpha\) in the numerator is the impedance-mismatch penalty. It is the same \(\alpha\) that appears in \(Z_{\rm orb} = Z_0\, n/\alpha\), which is the same \(\alpha\) that the Riemann sphere derivation of Section 1 identified as the departure from the golden-ratio winding attractor. The spectrometer, the circuit, and the Riemann sphere are all reading the same quantity.
The Norton equivalent circuit is the same boundary, read from the other side.
By Norton’s theorem, any two-terminal interface with a Thévenin equivalent \((V_{\rm TH},\, Z_{\rm orb})\) has a Norton equivalent consisting of a current source \(I_N\) in parallel with the same impedance \(Z_{\rm orb}\). The Norton current is the short-circuit current: the current that flows when the terminals are shorted (\(Z_{\rm load} = 0\)): \[I_N = \frac{V_{\rm TH}}{Z_{\rm orb}} = \frac{IE}{Z_0\, n/\alpha} = \frac{IE\,\alpha}{Z_0\, n}. \label{eq:norton_current}\] In the Norton circuit (right panel of Figure 4), \(Z_{\rm orb}\) appears as a shunt element — connected in parallel between the rails rather than in series in the top wire. The load \(Z_0\) is connected across the output terminals in the same way.
The two circuits are dual representations of the same physical boundary. The Thévenin representation reads the boundary from the E-sector: it is organized by the voltage \(V_{\rm TH} = IE\), the open-circuit electric potential that the nucleus maintains. The Norton representation reads the boundary from the H-sector: it is organized by the current \(I_N = IE/Z_{\rm orb}\), the short-circuit magnetic current that the orbital circulation generates.
The invariant of the duality is the power available at the boundary: \[V_{\rm TH} \cdot I_N = IE \cdot \frac{IE}{Z_{\rm orb}} = \frac{IE^2}{Z_{\rm orb}}. \label{eq:power_invariant}\] This product is the same regardless of which representation is used. It is the electromagnetic power available for transfer from the orbital to the vacuum — the power that becomes a photon when emission occurs.
The Thévenin circuit has a characteristic I–V curve: the straight line connecting the two intercepts.
The equation of the line is: \[I_{\rm out} = I_N - \frac{V_{\rm out}}{Z_{\rm orb}} = \frac{IE}{Z_{\rm orb}} - \frac{V_{\rm out}}{Z_{\rm orb}}. \label{eq:IV_line}\] Every point on this line is a possible operating point for the orbital–vacuum interface, depending on the load connected.
Now: where on this line does the impedance-match condition \(Z_{\rm orb} = Z_0\) correspond? This would require \(n/\alpha = 1\), i.e., \(n = \alpha \approx 1/137\) — not a positive integer. No individual orbital is impedance-matched to the vacuum. But there is a transition energy at which the field would momentarily pass through the impedance-match configuration. That energy is \(IE/4\).
On the I–V curve, \(G \approx IE/4\) corresponds to \(V_{\rm out} = IE/4 = V_{\rm TH}/4\): one quarter of the way from the origin to the x-intercept. This is not the maximum-power point (which is at \(V_{\rm TH}/2\), where load equals source impedance in a resistive circuit). It is the specific voltage at which the transition energy matches the impedance boundary of the vacuum.
For a symmetric ion — one where the spectral weight is equally distributed between the capacitive and inductive sectors around this boundary — the gap center \(G\) sits exactly at \(IE/4\) on the voltage axis. For an asymmetric ion, \(G\) is displaced along the same I–V line. The I–V line does not change between ions of the same \(n\); it is fixed by \(Z_{\rm orb}\) and \(IE\). What changes is where along that line the gap center falls.
Section 5 derives why \(IE/4\) is the impedance boundary and proves that no bound-to-bound transition can carry this energy. Section 4 introduces the \(\kappa\) coordinate that places this boundary at the same position in every ion’s spectrum, regardless of nuclear charge.
We have established that the orbital–vacuum boundary has a reflection coefficient \(\Gamma(n) \to 1\) and a transmission coefficient \(T(n) = 4\alpha n/(n+\alpha)^2 \approx 4\alpha/n\). The reflection is nearly total. But it is not total. A small fraction of the field amplitude crosses the boundary at every attempt. The question we now ask is: what exactly crosses, what does it carry, and why does it carry energy in units of \(h\nu\)?
The answer to all three questions comes from the geometry of the logarithmic field established in Section 1, applied to the mode conversion that occurs at the boundary. No new postulates are needed. Planck’s relation \(E = h\nu\) is a consequence of the self-similar structure of the vacuum’s field hierarchy.
Return to the logarithmic potential \(\phi(r) = \log(1/r)\) generated between the two poles of the Riemann sphere. The equipotential surfaces of this field are concentric circles at radii \(r\), \(re^{-\Delta}\), \(re^{-2\Delta}\), \(\ldots\) for any uniform depth step \(\Delta\). These surfaces nest geometrically: each is smaller than the previous by the constant factor \(e^{-\Delta}\).
As the process unfolds at rate \(c\), these nested surfaces propagate outward through space. A fixed observer at distance \(r\) from the source experiences a sequence of equipotential surfaces passing through their location. The rate at which these surfaces arrive is the frequency \(\nu\) of the field oscillation at that location. The spatial arrangement of nested equipotential surfaces, propagating at \(c\), is what Maxwell’s equations describe as electromagnetic radiation.
This is not yet a photon — it is the structure of the propagating \(Z_0\) field. The photon emerges when a local closure releases a bounded piece of this structure into free propagation. We now show what determines the energy of that piece.
The logarithmic field is self-similar: the geometry at any depth looks identical to the geometry at every other depth, scaled by a constant factor in \(r\). Section 1 established that the scaling ratio of this field converges uniquely to \(\varphi - \alpha\), and that the depth coordinate \(\kappa\) measures position in this hierarchy in units of \(\alpha\). At any depth \(\kappa\), the characteristic energy of the field is: \[E(\kappa) = E_0\,\alpha^{\kappa}, \label{eq:energy_ladder}\] where \(E_0 = m_e c^2 \alpha^2/2\) is the Rydberg energy — the energy scale at which the hierarchy is anchored. This is the energy ladder: each rung is one factor of \(\alpha\) below the previous, in exact correspondence with the logarithmic depth of the field.
The field is also self-similar in frequency. The oscillation frequency at depth \(\kappa\) scales identically: \[\nu(\kappa) = \nu_0\,\alpha^{\kappa}, \label{eq:frequency_ladder}\] where \(\nu_0\) is the empirically measured frequency intercept that anchors the ladder at the ionization scale. This is the Thread Law. It is important to be precise about its epistemic status, because the two scaling relations — equation [eq:energy_ladder] and equation [eq:frequency_ladder] — are independent results, not a single claim dressed twice.
The energy ladder \(E(\kappa) = E_0\,\alpha^\kappa\) is derived from NIST energy levels alone: the resonance sweep identifies level pairs whose spacing matches \(\Delta E_{\rm target}(\kappa;\,Z) = E_0 Z^2\hat{\mu}\,\alpha^\kappa\) within tolerance. No photon frequencies, no \(h\), and no \(\nu\) are involved at this step. The \(\kappa\) coordinate and the \(\alpha^\kappa\) structure of the energy hierarchy emerge purely from the Hamiltonian.
The frequency ladder \(\nu(\kappa) = \nu_0\,\alpha^\kappa\) is a downstream empirical discovery. The resonant level pairs already identified from energy levels are matched against NIST observed photon transitions. When those photon frequencies are plotted in the \(\kappa\) coordinate, their alignment on the slope \(\beta = \log_{10}\alpha\) is not built into the method and is not derivable from the coordinate definition. It is simply observed.
The ratio \(h = E_0/\nu_0\) is therefore the measured invariant of two independent \(\alpha^\kappa\) hierarchies that were found to agree: \(E_0\) from the energy-level sweep, \(\nu_0\) from the photon-frequency alignment. \(h\) is not assumed. It is what the agreement between the two hierarchies measures. The Thread Law is confirmed to residuals of \(9.47 \times 10^{-13}\) in hydrogen (K. B. Heaton and Coherence Research Collaboration 2026c).
Now form the ratio. At every depth \(\kappa\), the ratio of energy to frequency is: \[\frac{E(\kappa)}{\nu(\kappa)} = \frac{E_0\,\alpha^{\kappa}}{\nu_0\,\alpha^{\kappa}} = \frac{E_0}{\nu_0} = h. \label{eq:planck_from_selfsimilar}\] The factors of \(\alpha^\kappa\) cancel exactly. The ratio is independent of depth. It is the same at the ionization scale (\(\kappa = 0\)), at the impedance boundary (\(\kappa = \kappa_q \approx 0.282\)), at the fine-structure scale (\(\kappa = 2\)), and at every other depth simultaneously.
This is Planck’s constant \(h\). It is not a measured input to the theory. It is the scaling invariant of the self-similar logarithmic field: the quantity that cannot change because both energy and frequency scale identically with \(\alpha^\kappa\) at every depth. \(h\) is the action — energy times period, or energy divided by frequency — that the field preserves at every scale. Its constancy is the direct expression of what self-similarity means: a field that looks the same at every depth must perform the same action per cycle at every depth, and that action is \(h\).
A self-similar process has no preferred scale. If the field at depth \(\kappa\) is geometrically identical to the field at depth \(\kappa + 1\) (scaled by \(\alpha\)), then the ratio of any scale-dependent quantity to any other scale-dependent quantity must be the same at both depths. Energy is scale-dependent: \(E \propto \alpha^\kappa\). Frequency is scale-dependent: \(\nu \propto \alpha^\kappa\). Their ratio \(E/\nu\) is therefore scale-independent: it is the same at every depth, at every transition, in every atom.
That scale-independent ratio is \(h\). The self-similar geometry of the logarithmic field generates exactly one dimensional invariant of this kind, and that invariant is Planck’s constant. \(h\) is not the quantum of action in the sense of being a smallest allowed unit of action. It is the action of the self-similar process expressed once, at the scale of one complete cycle. The universe does not have a smallest action; it has a universal action — one that is the same at every scale because the process that generates it is self-similar at every scale.
When the transmission coefficient \(T(n) > 0\) allows a portion of the field to cross the orbital–vacuum boundary, a specific thing happens. We call it calving, by analogy with the calving of a glacier: a bounded piece of the local closure breaks free from the containing boundary and propagates independently.
The calved piece cannot carry the orbital field impedance \(Z_{\rm orb} = Z_0\, n/\alpha\). The vacuum does not support this impedance. The vacuum supports exactly one field configuration: a field with ratio \(|\mathbf{E}|/|\mathbf{H}| = Z_0\), propagating at \(c\). Whatever crosses the boundary must immediately reorganize into this configuration. This reorganization is not optional. It is a constraint imposed by what the vacuum is.
The result of the reorganization is a free-propagating field with:
The calved field is a photon. It satisfies Maxwell’s equations because it carries \(Z_0\) and propagates at \(c\) — precisely the conditions for a plane-wave solution of the vacuum field equations established in Section 2. It is not created from nothing. It is the \(Z_0\) relationship — the third element of the primitive structure — that was previously held in distorted form (\(Z_{\rm orb} \gg Z_0\)) within the orbital closure, now liberated to propagate in its natural, undistorted form.
From equation [eq:planck_from_selfsimilar], the ratio of energy to frequency is \(h\) at every depth of the self-similar hierarchy. The calved field carries energy \(\Delta E\) and is a piece of this same hierarchy — it has not left the hierarchy, it has merely changed from a bound mode to a free mode within it. Its frequency is therefore determined by the same invariant: \[\nu = \frac{\Delta E}{h}. \label{eq:planck_relation}\] This is Planck’s relation \(E = h\nu\), derived from the self-similar geometry of the logarithmic field. It is not a postulate of quantum mechanics. It is a consequence of three facts:
The photon therefore carries three things: the \(Z_0\) relationship (imposed by the vacuum), the energy \(\Delta E\) (imposed by energy conservation), and the frequency \(\nu = \Delta E/h\) (imposed by the self-similar scaling invariant \(h\) of the logarithmic field). None of these are postulated. All three are necessary consequences of the geometry.
The photon does not create new matter or new primitives; it creates new geometric relationships — new instances of the \(Z_0\) ratio propagating freely between what were previously separate local closures. The structural primitive \(\varepsilon_0\) and the non-structural primitive \(\mu_0\) are not created or destroyed by photon emission. What the photon creates — what it is — is a new instance of geometry: the \(Z_0\) relationship propagating freely, at rate \(c\), carrying the energy difference between two local closures.
Before emission, the orbital at quantum number \(n\) holds the \(Z_0\) relationship in distorted form. The electric and magnetic field amplitudes at the orbital radius are in ratio \(Z_{\rm orb} = Z_0\, n/\alpha \gg Z_0\). The relationship exists but is constrained. After emission, two things exist: a tighter local closure at \(n' < n\) (the \(Z_0\) relationship, more distorted, more deeply bound) and a free-propagating photon (the \(Z_0\) relationship, undistorted, liberated). The process is: \[\underbrace{Z_{\rm orb}(n)}_{\text{distorted, bound}} \;\longrightarrow\; \underbrace{Z_{\rm orb}(n')}_{\text{more distorted, more bound}} \;+\; \underbrace{Z_0\;\text{propagating}}_{\text{undistorted, free}}. \label{eq:calving_process}\] The total geometric content is conserved. The primitives are conserved. What changes is the distribution of geometric constraint: less constraint is held locally (the tighter orbital at \(n'\)), and the released constraint propagates freely as a photon.
When the photon is absorbed by another atom, the reverse occurs: the free-propagating \(Z_0\) relationship is incorporated into a new local closure, reorganizing that atom’s orbital structure upward. Geometry is redistributed. Matter is not created or destroyed. The universe is a field of propagating geometric relationships that continuously form and dissolve local closures, exchanging geometry between them.
The circuit picture of the orbital–vacuum interface is not an approximation to a more fundamental quantum description. It is the correct description at the level of field geometry. And it has an immediate consequence for the circuit model itself: no circuit is possible without electromagnetic radiation.
Every circuit element is a local closure of the \(Z_0\) relationship in some form. A resistor is a collection of atomic closures that continuously calve photons — heat radiation — because the impedance mismatch between the conduction electrons and the atomic lattice generates continuous mode conversion at the atomic scale. Ohm’s law \(V = IR\) is the macroscopic expression of the statistical average of trillions of microscopic calving events per second.
An inductor stores energy in the magnetic field — the open, non-structural mode \(\mu_0\). The energy stored is \(\tfrac{1}{2}LI^2 = \tfrac{1}{2}\mu_0 H^2 \times \text{volume}\): the non-structural primitive holding the field without yet calving it. When the inductor discharges, the magnetic energy converts to electric energy and eventually to radiation: the \(\mu_0\)-mode releases into the \(Z_0\) propagating mode.
A capacitor stores energy in the electric field — the contained, structural mode \(\varepsilon_0\). The energy stored is \(\tfrac{1}{2}CV^2 = \tfrac{1}{2}\varepsilon_0 E^2 \times \text{volume}\): the structural primitive holding the field without yet calving it.
In a lossless LC oscillator, the energy alternates between the two modes without net radiation. But no real circuit is lossless: every real circuit has resistance, every resistance dissipates power as heat, every caloric photon carries \(h\nu\) of energy away from the circuit. The photon is not an optional feature of circuit theory. It is the mechanism by which energy leaves any local closure and returns to the propagating field.
This is the circuit picture made complete. The Thévenin model describes the boundary between one local closure (the orbital) and the propagating field (the vacuum). The photon is what crosses that boundary in the direction from bound to free. Its energy is \(\Delta E\), its frequency is \(\Delta E/h\), its field ratio is \(Z_0\), its propagation obeys Maxwell’s equations. All four properties are derived from the geometry. None is postulated.
The calving process does not terminate. When an orbital at quantum number \(n\) calves a photon and reorganizes to \(n'\), the new orbital is a valid local closure with its own reflection coefficient \(\Gamma(n') \to 1\) and its own transmission coefficient \(T(n')\). The process can continue, level by level, until the atom reaches the ground state \(n = 1\): the deepest local closure, with \(Z_{\rm orb}(1) = Z_0/\alpha \approx 137\, Z_0\), the maximum impedance mismatch, the minimum transmission coefficient \(T(1) \approx 4\alpha \approx 0.029\).
From the ground state, no further calving to a lower orbital is possible. The atom awaits the next external perturbation — a photon absorbed from the propagating field, a collision, an external field — that raises it to a higher closure level and allows the calving process to resume.
The ground state is not a dead end. It is the atom’s deepest participation in the original encounter: the most tightly bound local closure that the Coulomb field can sustain under the constraint that \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) and \(n \geq 1\). From the ground state, the atom is most opaque to the vacuum, most deeply mismatched, most nearly a perfect resonant trap. And because \(T(1) > 0\) — because \(\alpha \neq 0\), because the crack in the structural primitive is real and non-zero — the ground state is not perfectly stable either. Even from \(n = 1\), there is a finite probability of interaction with the propagating field.
The fine-structure constant \(\alpha\) is therefore not merely the coupling constant of electromagnetism in the standard sense. It is the measure of the crack in the structural primitive that ensures the process is non-terminal at every level. If \(\alpha = 0\), the reflection coefficient would be exactly \(1\) at every orbital, the transmission coefficient would be exactly \(0\), and the local closures would be perfectly stable forever. No photon emission. No chemistry. No time. The process would exist but never unfold.
\(\alpha \neq 0\) is what makes the universe dynamic. It is the crack through which geometry propagates from the local closure to the propagating field and back. It is the measure of the original defect in the structural primitive, preserved at every scale of the self-similar hierarchy, expressed in every photon that has ever been emitted. The geometric calving described in this section operates at the atomic scale, where the relevant boundary is the orbital–vacuum interface and the relevant closure condition is the Coulomb field. But the same principle — that local closures redistribute \(Z_0\) relationships by calving geometry across impedance boundaries — applies at every scale at which closure conditions are satisfied. The question of how iterative calving builds nested closure structure across scales — and whether the known mass floors at the atomic, nuclear, and hadronic levels are the discrete rungs of a single geometric hierarchy that also governs the accumulation and dissolution of complex closure geometry — is taken up in the Discussion (Section 7).
Beginning from Coulomb’s law and the Biot–Savart law, applied to the electromagnetic closure mode at principal quantum number \(n\) with closure velocity \(v_n = \alpha c/n\):
Sections 1 through 3 have established the geometric foundation: two primitives, their encounter on the Riemann sphere, the logarithmic field with its \(\varphi\)-attractor perturbed by \(\alpha\), Maxwell’s equations as the interior description, the orbital impedance \(Z_{\rm orb}(n) = Z_0\, n/\alpha\), the Thévenin circuit, and the photon as the calved \(Z_0\) relationship.
We have not yet looked at a single spectrum.
Before we do, we need a ruler that is adapted for our purpose. Not a ruler calibrated to electron-volts — that is the ruler we already have, and it hides the structure we are looking for. We need a ruler calibrated to the vacuum itself: one whose tick marks are set by \(\alpha\), so that the logarithmic hierarchy of the Riemann sphere field is directly readable as a position on the scale 3. With this ruler, the spectroscopic consequence of the orbital–vacuum boundary — the gap at \(IE/4\) — appears at exactly the same position in every atom’s spectrum, regardless of nuclear charge, electron configuration, or ionization stage.
This section defines that ruler, motivates it from the geometry, and shows that it places the impedance boundary at a fixed universal coordinate. The next section uses it to derive and confirm the gap.
Consider two ions: chromium I (Cr i) with ionization energy \(IE = 6.767\) eV and cerium II (Ce ii) with \(IE = 10.850\) eV. The impedance boundary \(IE/4\) falls at \(1.692\) eV for Cr i and at \(2.713\) eV for Ce ii. These are different numbers with no obvious connection. In absolute electron-volts, there is no reason to expect a relationship between the spectral structures of two such different ions. The nuclear charges differ by a factor of nearly three. The electron configurations are completely different. The ionization energies differ by sixty percent.
And yet: the impedance boundary is the same condition in both ions — the energy at which \(Z_{\rm orb} = Z_0\), where the bound mode would momentarily achieve impedance match with the vacuum. The condition is universal; its expression in electron-volts is not.
To see the universal structure, we need to express every transition energy as a fraction of the ionization energy, measured in units of \(\alpha\). The result is a coordinate in which Cr i and Ce ii — and every other ion in the periodic table — place their impedance boundaries at exactly the same position.
The Riemann sphere’s logarithmic field generates a natural energy hierarchy: each rung is a factor of \(\alpha\) below the previous, anchored at the ionization energy of the ion. The depth of any transition energy \(\Delta E\) in this hierarchy — measured from the ionization limit — is the coordinate \(\kappa\).
Definition 1 (\(\kappa\) coordinate). For any transition energy \(\Delta E\) in an ion of atomic number \(Z\) and first ionization energy \(IE\), the hydrogenic depth coordinate is: \[\kappa(\Delta E;\, Z) = \frac{\log_{10}(\Delta E\, /\, E_0 Z^2 \hat{\mu})}{\log_{10} \alpha}, \label{eq:kappa_definition}\] where \(E_0 = m_e c^2\alpha^2/2 \approx 13.606\) eV is the Rydberg energy (infinite nuclear mass limit), and \(\hat{\mu} = m_{\rm nuc}/(m_e + m_{\rm nuc})\) is the reduced-mass correction factor for the ion (tabulated per ion; \(\hat{\mu} \to 1\) for heavy nuclei).
Every symbol of the “Thread Frame” coordinate system, explained:
|
\(\kappa\) |
Energy scale |
Physical significance |
|
\(0\) |
\(E_0 Z^2\hat{\mu}\) |
ionization limit |
|
\(\kappa_q \approx 0.282\) |
\(IE/4\) |
Impedance boundary (this paper) |
|
\(1\) |
\(E_0 Z^2\hat{\mu} \cdot \alpha\) |
One \(\alpha\)-step below ionization |
|
\(2\) |
\(E_0 Z^2\hat{\mu} \cdot \alpha^2\) |
Fine-structure scale |
The fine-structure constant was historically measured at \(\kappa = 2\), because that is the depth at which \(\alpha^2\) corrections to the Bohr energies first become spectroscopically visible. This is why \(\alpha\) was called the “fine-structure” constant: it was first seen at the fine-structure depth. On the \(\kappa\) ruler, \(\kappa = 2\) is the second tick mark down — not the most significant feature of the spectrum, but the first one where \(\alpha\) was historically legible. The most significant feature lies between \(\kappa = 0\) and \(\kappa = 1\), at \(\kappa_q \approx 0.282\), and it had not been identified until this work.
The impedance boundary \(\Delta E = IE/4\) has a definite position on the \(\kappa\) ruler. We compute it now.
By definition of \(\kappa\): \[\begin{aligned} \kappa(IE/4;\, Z) &= \frac{\log_{10}\!\bigl((IE/4)\,/\,E_0 Z^2\hat{\mu}\bigr)} {\log_{10} \alpha} \nonumber \\[6pt] &= \frac{\log_{10}(1/4)}{\log_{10} \alpha} \nonumber \\[6pt] &= \frac{\log_{10} 4}{|\log_{10} \alpha|} \label{eq:kappa_q_derivation} \end{aligned}\] where the second step uses \(IE = E_0 Z^2\hat{\mu}\) (the ionization energy equals the hydrogenic energy scale for each ion, exactly by definition), so the \(Z\)-dependence cancels. The result is: \[\kappa_q = \frac{\log_{10} 4}{|\log_{10} \alpha|} \approx \frac{0.60206}{2.13685} \approx 0.2818. \label{eq:kappa_q}\]
This number, \(\kappa_q \approx 0.2818\), is universal. It does not depend on \(Z\). It does not depend on the electron configuration. It does not depend on the ionization stage. For every ion in the periodic table, for every atom ever observed, the impedance boundary \(IE/4\) sits at depth \(\kappa_q \approx 0.2818\) on the \(\kappa\) ruler. The \(Z\)-dependence cancels exactly because \(Z_{\rm orb}/Z_0 = n/\alpha\) does not depend on \(Z\) — the nuclear charge enters through \(a_0 \propto 1/Z\) and \(r_n \propto n^2/Z\), and these cancel in the ratio \(|\mathbf{E}_n|/|\mathbf{H}_n|\) exactly as we showed in Section 3.4.
The orbital impedance is \(Z_{\rm orb}(n) = Z_0\, n/\alpha\). This result was derived in Section 3.4 from Coulomb’s law and the Biot–Savart law. During the derivation, the orbital radius \(r_n = n^2 a_0\) and the elementary charge \(e\) both cancelled from the ratio \(|\mathbf{E}_n|/|\mathbf{H}_n|\). For a multi-electron ion, the orbital radius is modified by screening: \(r_n \approx n^2 a_0/Z_{\rm eff}\). But this modification cancels in the same way — \(r_n\) appears in both \(|\mathbf{E}_n|\) and \(|\mathbf{H}_n|\) and disappears from their ratio. The impedance \(Z_{\rm orb} = Z_0\, n/\alpha\) is independent of \(Z\) at every principal quantum number. The impedance boundary \(Z_{\rm orb} = Z_0\) is therefore the same condition in every ion, expressed at the energy \(IE/4\) in every ion’s own energy units.
This is why the gap is universal. It is not a property of hydrogen, or of the d-block, or of any particular element. It is a property of the vacuum impedance \(Z_0\), imposed on every Coulomb field that closes into a bound orbital mode, at the one energy where the bound and free field configurations would exchange.
The \(\kappa\) coordinate reveals a second universal feature of atomic spectra: the frequency backbone.
In Section 3.10, we established that both the energy ladder and the frequency ladder of the self-similar logarithmic field scale as \(\alpha^\kappa\): \[\nu(\kappa) = \nu_0\, \alpha^\kappa. \label{eq:thread_law}\] 4 Taking \(\log_{10}\) of both sides: \[\log_{10}\nu(\kappa) = \log_{10}\nu_0 + \kappa\,\log_{10}\alpha = \chi + \beta\,\kappa, \label{eq:thread_law_log}\] where \(\chi = \log_{10}\nu_0\) is the ion-specific intercept (depending on the ionization energy) and: \[\beta = \log_{10}\alpha \approx -2.1368. \label{eq:beta}\]
This is the Thread Law: the prediction that photon frequencies, plotted against depth \(\kappa\) on a logarithmic frequency axis, organize onto a straight line with universal slope \(\beta = \log_{10}\alpha\).
The slope \(\beta\) is not fitted to any spectrum. It is computed from \(\alpha\) alone. Its physical meaning is the rate at which photon frequency decreases per unit of \(\kappa\)-depth: each step deeper into the bound spectrum multiplies the photon frequency by \(\alpha\). This is the self-similar logarithmic field on the Riemann sphere, made directly visible in the photon frequencies of any atom.
For hydrogen, the Thread Law can be checked with no free parameters. Using NIST spectral data for the analytic hydrogen template (levels \(E_n = E_0/n^2\), all transitions with \(n_f < n_i \leq 60\)):
The agreement at twelve significant figures is not an accident of fitting; the slope is \(\log_{10}\alpha\) and the intercept is set by the Rydberg energy. The residual of \(9.47 \times 10^{-13}\) is the measurement precision of the NIST frequency data, not a model residual.5 Thus the logarithmic hierarchy of the Riemann sphere field is not a theoretical construct, but a directly measurable feature of the hydrogen spectrum, confirmed to the precision of the best available spectroscopic data.
It is fair to ask: if the \(\kappa\) coordinate is defined using \(\alpha^\kappa\), why is it surprising that \(\nu\) scales with \(\alpha^\kappa\)? The coordinate is easy to define. What is not guaranteed is that nature populates it. Most level pairs in any ion’s NIST catalog do not land near \(\Delta E_{\rm target}(\kappa;\, Z)\). The empirical content of the Thread Law is threefold: a statistically significant excess of resonant level pairs occurs at specific \(\kappa\) bins, beyond what is expected from the level-density structure alone; a non-trivial subset of those aligned pairs correspond to measured photon transitions that match within spectroscopic uncertainty windows; and when those photons are plotted in \((\kappa,\, \log_{10}\nu)\) space they form coherent threads with the predicted slope \(\beta = \log_{10}\alpha\), with small structured deviations that carry additional physical information (the “intercepts,” \(\chi\)). The coordinate does not force the data onto the line. The data lands on the line because the self-similar geometry of the vacuum field is what organizes atomic spectra. The ion-specific intercepts \(\chi\) are not arbitrary offsets: they encode the electromagnetic character of each ion — its E/H balance at the impedance boundary, the same asymmetry that displaces the gap center \(G\) from \(IE/4\) in electromagnetically asymmetric ions (Section 5.6). The geometric relationship between \(\chi\) and the E/H partition is an open frontier of this research program.
The \(\kappa\) ruler reveals something that was invisible in electron-volts: the slope \(\beta = \log_{10}\alpha\) of the photon frequency backbone and the slope \(-1/Z_{\rm orb}\) of the Thévenin I–V characteristic are not merely analogous. They are the same quantity \(\alpha\), expressed in two different domains.
The I–V slope is: \[-\frac{1}{Z_{\rm orb}(n)} = -\frac{\alpha}{Z_0\, n}. \label{eq:IV_slope}\] The \(\beta\) slope, expressed in frequency-per-\(\kappa\)-unit, gives a factor of \(\alpha\) per unit depth. In consistent units: \[-\frac{1}{Z_{\rm orb}(n)} = \frac{10^\beta}{Z_0\, n}. \label{eq:beta_zorb_identity}\] since \(10^\beta = 10^{\log_{10}\alpha} = \alpha\). This is an exact identity, not an approximation. The I–V slope and the \(\beta\) slope share \(\alpha\) as their common factor and are related by: \[\text{I--V slope} = -\frac{1}{Z_{\rm orb}} = \frac{10^\beta}{Z_0\, n}. \label{eq:two_slopes_unified}\]
The spectrometer reads \(\beta\): the rate of change of \(\log\nu\) per unit of \(\kappa\)-depth, which is \(\log_{10}\alpha\). The circuit reads \(-1/Z_{\rm orb}\): the rate of change of current per unit of voltage at the orbital boundary, which is \(-\alpha/(Z_0 n)\). Both are reading the same quantity — the departure of the actual winding ratio from the \(\varphi\)-attractor, which the Riemann sphere geometry forces to equal \(\alpha\). The spectrometer and the circuit model are not two different theories. They are two instruments aimed at the same geometry.
|
Domain |
Quantity |
\(\alpha\) appears as |
|
Riemann sphere |
\(\xi_{\rm actual} = \varphi - \alpha\) |
departure from \(\varphi\)-attractor |
|
Spectral / logarithmic |
\(\beta = \log_{10}\alpha\) |
slope of frequency backbone |
|
Circuit / admittance |
\(-1/Z_{\rm orb} = -\alpha/(Z_0 n)\) |
slope of I–V characteristic |
|
Estimator / geometry |
\(G = \sqrt{{\rm gap_{lo}} \cdot {\rm gap_{hi}}}\) |
midpoint in \(\alpha\)-scaled log space |
Section 4.5 established an exact identity: the spectral slope \(\beta\) and the circuit admittance slope \(-1/Z_{\rm orb}\) are the same quantity \(\alpha\) expressed in two different domains. The identity is correct and important. But it does not yet answer the deeper question: why are they the same? Two instruments reading the same quantity must be reading a common source. This section identifies that source.
It is the Laplace equation — applied not to the potential directly, but to the logarithm of the photon frequency.
Section 1 established that the electromagnetic vacuum generates a logarithmic potential field between the two poles of the Riemann sphere: \[\phi(r) \;=\; \log\!\left(\frac{1}{r}\right) \;=\; -\log r, \qquad \nabla^2\phi \;=\; 0 \label{eq:log_potential_laplace}\] where \(r\) is the radial coordinate on the sphere (\(r \to 0\) at the source pole \(p\), \(r \to 1\) at the equator). This is the Green’s function of the Laplacian in two dimensions: the unique rotationally symmetric solution with a point source at \(p\).
Section 3.10 established the energy-potential coupling: the energy at depth \(\kappa\) is \(E(\kappa) = E_0\,\alpha^\kappa\). In terms of \(\phi\), since \(r = \alpha^\kappa\) and \(\phi = -\log r\): \[E \;=\; E_0\, r \;=\; E_0\, e^{-\phi}. \label{eq:energy_potential_coupling}\] Energy is proportional to the radial coordinate — or equivalently, it falls exponentially with the potential depth. This is not an additional assumption. It is the restatement of \(E(\kappa) = E_0\,\alpha^\kappa\) in the language of the field \(\phi = \kappa|\ln\alpha|\). The coupling \(E = E_0\,e^{-\phi}\) is the energy ladder of Section 3.10, written in field coordinates.
Planck’s relation \(\nu = E/h\) (Theorem 37 of (Coherence Research Collaboration 2026), derived as the scaling invariant in Section 3.10) then gives: \[\nu \;=\; \frac{E_0}{h}\, e^{-\phi}, \qquad \log\nu \;=\; \underbrace{\log\!\left(\frac{E_0}{h}\right)}_{\chi} \;-\; \phi. \label{eq:log_nu_phi}\] The log-frequency is the potential \(\phi\), negated and shifted by a constant. Since \(\phi\) satisfies Laplace’s equation, so does \(\log\nu\): \[\boxed{\nabla^2(\log\nu) \;=\; 0.} \label{eq:laplace_log_nu}\] This is the governing equation of photon frequency in the vacuum geometry. We call it the Harmonic Light Law.6 Both the spectrometer and the circuit are reading it.
The unique spherically symmetric solution to \(\nabla^2(\log\nu) = 0\) on \(S^2\), with a point source at \(p\) and bounded behavior at \(p^*\), is \(\log\nu \propto -\phi\) (the Green’s function of the Laplacian). In the \(\kappa\) coordinate, \(\phi = \kappa |\ln\alpha|\), so: \[\log_{10}\nu(\kappa) \;=\; \chi + \beta\kappa, \qquad \beta = \log_{10}\alpha. \label{eq:thread_law_derived}\] This is the Thread Law (Section 4.4). It is not an empirical pattern that the geometry happens to fit. It is the unique solution consistent with the field’s boundary conditions on the sphere. Any other slope would violate the Laplace equation for \(\log\nu\).
Since \(V = E = h\nu\) (the Thévenin voltage equals the transition energy, which equals Planck’s relation), \(\log V = \log h + \log \nu\), and therefore: \[\nabla^2(\log V) \;=\; \nabla^2(\log\nu) \;=\; 0. \label{eq:laplace_log_V}\] In the Thévenin circuit: \(V = IE\,e^{-\phi} = IE\,\alpha^\kappa\) (the voltage at depth \(\kappa\)) and \(I = (IE - V)/Z_{\rm orb}\) (Section 3.6). Then: \[\frac{dI}{dV} \;=\; \frac{dI/d\phi}{dV/d\phi} \;=\; \frac{V/Z_{\rm orb}}{-V} \;=\; -\frac{1}{Z_{\rm orb}}. \label{eq:IV_from_field}\] The I–V slope is the \(\phi\)-derivative ratio of the two circuit quantities. It is the circuit’s reading of the same field \(\phi\) that the spectrometer reads as \(\log\nu\).
The three-circuit progression of Figure 8 established the impedance scaffolding of the vacuum: Circuit 1 reads the orbital field through \(Z_{\rm orb}\), Circuit 2 reveals the mass floor \(Z_m\) at the ionization boundary, and Circuit 3 identifies \(Z_m\) as the Lorentz-scalar fiber-constant of the four-momentum. Both the spectrometer and the circuit are instruments reading a common source — but that source has not yet been named. The name is \(\nabla^2(\log\nu) = 0\), the Harmonic Light Law: the single field equation from which the spectral slope \(\beta = \log_{10}\alpha\) and the admittance \(-1/Z_{\rm orb}\) both descend as projections onto their respective domains (Figure 5).
The Harmonic Light Law \(\nabla^2(\log\nu)=0\) and its two instruments. The central node names the governing equation of photon frequency in the vacuum geometry, together with the energy-potential coupling \(E = E_0\,e^{-\phi}\) and Planck’s relation \(\nu = E/h\) from which it derives. Two instruments read this equation in distinct domains. The spectrometer reads \(\phi\) as the logarithm of photon frequency, yielding the Thread Law \(\log_{10}\nu = \chi + \beta\kappa\) with universal slope \(\beta = \log_{10}\alpha\). The Thévenin circuit reads \(\phi\) as the depth-dependent voltage \(V = IE\,e^{-\phi}\), yielding the I–V admittance \(dI/dV = -1/Z_{\rm orb} = -\alpha/(Z_0\,n)\). The brace identifies their common factor \(\alpha\): the two readings are the same quantity expressed in two domains. The impedance scaffolding — \(Z_{\rm orb}\), \(Z_m\), and the Lorentz-scalar mass floor — from which both instruments derive is shown in Figure 8.
Section 4.4 reported that the MDL analysis of (Kelly B. Heaton and Coherence Research Collaboration 2025a) confirmed the slope \(\beta = \log_{10}\alpha\) not only in atomic Rydberg spectra but across six independent spectral domains, including the solar spectrum and molecular C\(_2\) bands. This universality was confirmed empirically. It now has a theoretical explanation.
Every photon is the \(Z_0\) relationship propagating freely from a local closure (Section 2.4). For every photon, \(E = h\nu\) holds exactly. And the field \(\phi\) that the photon was emitted from satisfies \(\nabla^2\phi = 0\) by the Laplace equation of Section 1. Therefore every photon’s log-frequency satisfies \(\nabla^2(\log\nu) = 0\), regardless of which atom or molecule emitted it. The Thread Law’s slope \(\beta = \log_{10}\alpha\) is not a spectroscopic property of atomic Rydberg series. It is a topological property of the Laplace equation on the sphere. It cannot be otherwise for any photon propagating in the vacuum geometry established in Section 1.
The MDL confirmation is therefore not evidence that the theory happens to fit the data. It is confirmation that photon frequencies in nature are organized as a harmonic function — which is what the geometry requires them to be.
Since \(\log\nu\) is harmonic, the Cauchy-Riemann theorem guarantees that it is the real part of a holomorphic function on the sphere. By inspection — \(\log\nu = \mathrm{const} + \log r = \mathrm{const} + \log|z|\) — that function is the complex logarithm of the field coordinate. We call this the Complex Light Law: \[w \;=\; \log z \;=\; \underbrace{\log|z|}_{\displaystyle\mathrm{Re}(w)\,=\,\log\nu\big|_{\rm norm}} \;+\; i\underbrace{\arg(z)}_{\displaystyle\mathrm{Im}(w)\,=\,\theta}. \label{eq:complex_light_law}\] The Harmonic Light Law \(\nabla^2(\log\nu) = 0\) is the real-part expression of the Complex Light Law: it governs what the spectrometer reads. The imaginary part \(\theta = \mathrm{Im}(w) = \arg(z)\) governs what the H-sector instrument reads — the angular structure of the vacuum field, the subject of a companion paper in preparation. The real part is the normalised log-frequency, the imaginary part is \(\theta = \arg(z)\) — the winding phase around the source pole \(p\), the angular coordinate on the sphere. The two are related by the Cauchy-Riemann equations and are not independent: knowing \(\log\nu\) at every point of the sphere completely determines \(\theta\), and vice versa.
The Thread Law \(\nu \propto \alpha^\kappa\) is therefore the statement that photon frequency is proportional to the modulus of the complex field coordinate: \(\nu \propto |z|\). This has been the content of the Thread Frame since its introduction in (Kelly B. Heaton and Coherence Research Collaboration 2025b) — the original discovery \(\nu(\kappa) \propto \alpha^\kappa\) was the discovery of \(\nu \propto |z|\). It can now be stated as such.
The E/H partition of Section 2.2 is, at this level, the real/imaginary split of \(\log z\): the structural primitive \(\varepsilon_0\) organizes the radial mode \(|z|\) (frequency, energy, the Thread Frame), and the non-structural primitive \(\mu_0\) organizes the angular mode \(\arg(z)\) (phase, winding, the H-sector). This is why Ho ii’s spectral mass, concentrated in the angular sector, is invisible to the E-sector Thread Frame (Section 5.8): the Thread Frame reads \(\mathrm{Re}(\log z)\) and \(\mathrm{Im}(\log z)\) carries no projection onto it. The H-sector instrument reads \(\mathrm{Im}(\log z) = \theta\). Together they reconstruct \(z = |z|\,e^{i\theta}\) — the complete complex coordinate of the vacuum field.
The logarithmic field \(\phi\) satisfies Laplace’s equation. The energy is \(E = E_0\,e^{-\phi}\). Planck’s relation gives \(\nu = E/h\). Therefore \(\log\nu = \text{const} - \phi\), and since \(\phi\) is harmonic, so is \(\log\nu\).
The spectrometer reads \(\phi\) as a logarithm: it reports \(\log\nu = \chi + \beta\kappa\), a straight line whose slope is \(\beta = \log_{10}\alpha\).
The circuit reads \(\phi\) as an exponential: it reports \(V = IE\,e^{-\phi}\) and \(dI/dV = -1/Z_{\rm orb}\), a straight line in \((V, I)\) space whose slope is \(-\alpha/(Z_0 n)\).
One field. Two meters. One law: \(\nabla^2(\log\nu) = 0\).
More completely, the Complex Light Law: \(w = \log z = \log\nu + i\,\theta\). The Harmonic Light Law is its real part. The spectrometer reads \(\mathrm{Re}(w) = \log\nu\), measuring \(\nu \propto |z|\). The H-sector instrument reads \(\mathrm{Im}(w) = \theta\). Together they reconstruct \(z\) — the complete complex coordinate of the vacuum field.
The gap at \(IE/4\) has two measurable boundaries: the lower edge \({\rm gap_{lo}}\) (the last transition energy below \(IE/4\)) and the upper edge \({\rm gap_{hi}}\) (the first transition energy above \(IE/4\)). Their midpoint is the estimator \(G\) of the impedance boundary. What kind of midpoint?
The \(\kappa\) coordinate is logarithmic: equal steps in \(\kappa\) correspond to equal multiplicative factors in energy. In a logarithmic coordinate, the midpoint between two values is not their arithmetic mean but their geometric mean: \[G = \sqrt{{\rm gap_{lo}} \cdot {\rm gap_{hi}}}. \label{eq:geometric_mean}\] This is not an arbitrary choice of estimator. It is forced by the coordinate system. In \(\kappa\)-space, the midpoint between \(\kappa({\rm gap_{lo}})\) and \(\kappa({\rm gap_{hi}})\) is: \[\kappa_{\rm mid} = \frac{\kappa({\rm gap_{lo}}) + \kappa({\rm gap_{hi}})}{2} = \frac{1}{2}\!\left[ \frac{\log_{10}({\rm gap_{lo}}/E_0 Z^2\hat{\mu})} {\log_{10}\alpha} + \frac{\log_{10}({\rm gap_{hi}}/E_0 Z^2\hat{\mu})} {\log_{10}\alpha} \right]. \label{eq:kappa_midpoint}\] Converting back to energy units: \[E_{\rm mid} = E_0 Z^2\hat{\mu} \cdot \alpha^{\kappa_{\rm mid}} = \sqrt{{\rm gap_{lo}} \cdot {\rm gap_{hi}}} = G. \label{eq:G_from_kappa}\] The geometric mean is the midpoint in \(\kappa\)-space. An arithmetic mean \((\text{gap}_{\rm lo} + \text{gap}_{\rm hi})/2\) would be the midpoint in linear energy space, which is not the natural coordinate of the logarithmic field. Using the arithmetic mean would introduce a systematic bias in the estimation of \(IE/4\) that grows with gap width. The geometric mean is unbiased by construction.
There is a deeper reason why the geometric mean is the right estimator that goes beyond the logarithmic coordinate.
The gap has two boundaries with physically distinct meanings. The lower edge \({\rm gap_{lo}}\) is the terminus of the capacitive (E-sector) transitions: energies organized by the structural primitive \(\varepsilon_0\), where the electric mode dominates. The upper edge \({\rm gap_{hi}}\) is the onset of the inductive (H-sector) transitions: energies organized by the non-structural primitive \(\mu_0\), where the magnetic mode becomes accessible.
Their product \(G^2 = {\rm gap_{lo}} \cdot {\rm gap_{hi}}\) fixes one combination of the two sector boundaries; their ratio \({\rm gap_{hi}}/ {\rm gap_{lo}}\) fixes the other. The geometric mean \(G\) is the balance point at which neither sector dominates — which is exactly the impedance-match condition \(Z_{\rm orb} = Z_0\). The estimator has the form it has because it encodes the same two-invariant structure as the vacuum itself:
\[Z_0 = \sqrt{\frac{\mu_0}{\varepsilon_0}}, \qquad G = \sqrt{{\rm gap_{lo}} \cdot {\rm gap_{hi}}}. \label{eq:Z0_G_parallel}\]
Both are geometric means of two complementary sector quantities. \(Z_0\) is the geometric mean of the inductive and capacitive vacuum constants. \(G\) is the geometric mean of the inductive and capacitive sector boundaries in the spectrum. The structural parallel is exact.
We established in Section 3.10 that \(h\) is the scaling invariant of the self-similar logarithmic field. We established in Section 1 that \(\alpha = Z_0/2R_K\) is the departure from the \(\varphi\)-attractor. These two derivations are not independent: they are connected by a chain of exact identities that locks all four fundamental atomic constants together.
The four constants \(\alpha\), \(h\), \(m_e\), and \(a_0\) are related by: \[\begin{aligned} \alpha &= \frac{Z_0}{2R_K} = \frac{Z_0 e^2}{2h} \label{eq:alpha_lock} \\[6pt] E_0 &= \frac{m_e c^2 \alpha^2}{2} \label{eq:rydberg_lock} \\[6pt] a_0 &= \frac{\hbar}{m_e \alpha c} = \frac{\hbar c\,\alpha}{m_e c^2 \alpha^2} = \frac{\hbar c}{E_0}\cdot\frac{\alpha}{2} \label{eq:bohr_lock} \end{aligned}\] These are exact identities with no free parameters. Each one follows from the definitions of the constants involved.
The consequence is a logical constraint: if the geometry correctly derives \(\alpha\), and if the self-similar scaling invariant correctly gives \(h\), then \(m_e\) and \(a_0\) are not free to take any values — they are determined. Specifically, from the gap center \(G \approx IE/4 \approx E_0/4\) (in the infinite-nuclear-mass limit for hydrogen): \[m_e c^2 = \frac{8G}{\alpha^2}, \qquad a_0 = \frac{\hbar c\,\alpha}{8G}. \label{eq:mass_bohr_from_gap}\] These are the same equations established in the companion empirical paper (K. B. Heaton and Coherence Research Collaboration 2026c), derived there from the impedance-match condition. Here we see them as the algebraic consequence of the four-constant lock: if \(\alpha\) is correct and \(G = E_0/4\) is correct, then \(m_e\) and \(a_0\) follow with no freedom.
The model does not make four separate predictions. It makes one prediction — the geometry is correct — which expresses itself in four measurable constants that must all be simultaneously consistent. If any one is wrong, all are wrong. The verification pass at the end of this paper will check all four.
We now have everything we need to derive the gap. The tools, assembled in order:
The gap is what happens when a geometric requirement and a number-theoretic impossibility occupy the same location in the spectrum. It is not a line shifted or broadened. It is a literal absence: a zero-observation interval at a predictable location in every ion’s transition catalog. This section derives why it must exist, proves why it cannot be avoided, and presents the empirical confirmation across 71 ions with zero free parameters and zero falsifications.
The impedance-match condition \(Z_{\rm orb}(n) = Z_0\) requires: \[\frac{Z_0\, n}{\alpha} = Z_0 \quad \Longrightarrow \quad n = \alpha \approx \frac{1}{137}. \label{eq:match_condition}\] No positive integer satisfies \(n = \alpha\). The impedance match is never achieved at any individual orbital. This is the direct electromagnetic statement of what we established in Section 1: the orbital is always deeply mismatched to the vacuum, because \(\alpha \ll 1\) ensures \(Z_{\rm orb} \gg Z_0\) at every integer \(n\).
But the impedance match is a condition on transition energies, not on individual orbitals. A transition that carries the field from one closure configuration to another passes through the space of field configurations between them. During that passage, the field impedance moves continuously. There is a specific transition energy at which the field impedance would momentarily equal \(Z_0\) — the impedance-match point — as the field converts from the bound mode to the free mode.
Which transition energy? The answer follows from the structure of the hydrogen energy ladder. The ionization energy of the \(n = 2\) shell is: \[E_{n=2} = \frac{E_0}{4} = \frac{IE}{4}, \label{eq:n2_energy}\] where \(IE = E_0\) for hydrogen in the infinite-nuclear-mass limit. This is the energy at which the field, transitioning from a bound closure mode, reaches the \(n = 2\) continuum threshold — the lowest energy at which the inductive vacuum mode (the H-sector, organized by \(\mu_0\)) first becomes accessible, because \(n = 2\) is the first shell supporting \(\ell \geq 1\) orbital angular momentum.
Below \(IE/4\): transitions terminate on shells \(n_f \geq 2\), which support both \(\ell = 0\) and \(\ell \geq 1\) states. The electric mode dominates. The field is in the capacitive sector, organized by \(\varepsilon_0\), the structural primitive.
Above \(IE/4\): transitions terminate on \(n_f = 1\) only, the purely \(\ell = 0\) ground state. The magnetic mode becomes the organizing principle. The field is in the inductive sector, organized by \(\mu_0\), the non-structural primitive.
At \(IE/4\) itself: neither mode is stable. The field is at the impedance-match point, simultaneously required by the geometry to pass through \(Z_{\rm orb} = Z_0\) and, as the next subsection proves, unable to do so through any bound-to-bound transition.
The hydrogen analysis above identifies \(IE/4\) as the impedance boundary via the specific physics of the \(n = 2\) shell: it is the threshold at which \(\ell \geq 1\) orbital character first becomes energetically accessible, and therefore the lowest energy at which the H-sector mode can participate. This reasoning is hydrogen-specific. The universality of \(IE/4\) as the impedance boundary in every ion follows instead from the \(Z\)-independence of \(Z_{\rm orb}/Z_0 = n/\alpha\), established in Section 3.4: because the impedance ratio carries no dependence on nuclear charge, the impedance-match condition \(Z_{\rm orb} = Z_0\) falls at exactly \(IE/4\) in every ion’s own energy units regardless of electron configuration or subshell filling. The sector-boundary interpretation follows: whatever force law implements the closure, \(IE/4\) is the energy at which the bound geometry would momentarily achieve vacuum match — and therefore the energy that separates the capacitive sector below from the inductive sector above, universally. Section 5.2 proves that no bound-to-bound transition can carry this energy in any ion.
No bound-to-bound transition can carry energy \(IE/4\). This is not a selection rule. It is a statement about integers.
A bound-to-bound transition in hydrogen carries energy: \[\Delta E(n_f, n_i) = E_0\!\left(\frac{1}{n_f^2} - \frac{1}{n_i^2}\right), \label{eq:transition_energy}\] where \(n_f < n_i\) are positive integers. For this transition energy to equal \(IE/4 = E_0/4\), we would need: \[\frac{1}{n_f^2} - \frac{1}{n_i^2} = \frac{1}{4}. \label{eq:diophantine}\] We now prove exhaustively that equation [eq:diophantine] has no solution in positive integers \(n_f < n_i\).
Equation [eq:diophantine] requires \(1 - 1/n_i^2 = 1/4\), giving \(1/n_i^2 = 3/4\), so \(n_i^2 = 4/3\). Not an integer.
Equation [eq:diophantine] requires \(1/4 - 1/n_i^2 = 1/4\), giving \(1/n_i^2 = 0\), so \(n_i \to \infty\). Not a bound state.
The maximum possible transition energy is the series limit \(\Delta E_{\rm max}(n_f) = E_0/n_f^2\), achieved as \(n_i \to \infty\). For \(n_f \geq 3\): \[\Delta E_{\rm max}(n_f) = \frac{E_0}{n_f^2} \leq \frac{E_0}{9} < \frac{E_0}{4} = \frac{IE}{4}. \label{eq:max_energy_case3}\] No transition terminating on \(n_f \geq 3\) can carry enough energy to reach \(IE/4\). \(\square\)
The three cases are exhaustive. The gap at \(IE/4\) is therefore not a feature of a particular atomic model, not a selection rule that a sufficiently exotic transition could violate, and not an approximation that breaks down at high energies. It is a number-theoretic identity: the equation \(1/n_f^2 - 1/n_i^2 = 1/4\) has no solution in positive integers, and it never will. The gap is permanent, universal, and exact for hydrogen.
For multi-electron ions, the Diophantine argument extends by a three-step chain.
First, \(Z_{\rm orb}/Z_0 = n/\alpha\) is independent of nuclear charge \(Z\). This was established in Section 3.4: the elementary charge \(e\) and the orbital radius \(r_n\) both appear in the numerator and denominator of \(|\mathbf{E}_n|/|\mathbf{H}_n|\) and cancel exactly. The impedance ratio depends only on \(n\) and \(\alpha\), not on \(Z\).
Second, screening modifies the effective orbital radius \(r_n\) in multi-electron ions, but the cancellation is preserved. The ratio \(|\mathbf{E}|/|\mathbf{H}|\) at the orbital radius is independent of \(Z_{\rm eff}\) for the same reason it is independent of \(Z\) in hydrogen: the radius appears in both field components and cancels from the ratio (as noted in the boxed summary of Section 4.3).
Third, therefore \(IE/4\) is the impedance boundary in every ion’s own energy units, regardless of \(Z\), screening, or subshell configuration. The Diophantine impossibility holds in every ion because the impedance condition \(Z_{\rm orb} = Z_0\) falls at \(IE/4\) in every ion’s spectrum by the same algebraic argument, and the equation \(1/n_f^2 - 1/n_i^2 = 1/4\) remains unsatisfied in positive integers regardless of the ion that instantiates the shell structure.
The argument is geometric, not arithmetic, and it holds for every multi-electron ion: a field cannot simultaneously satisfy the closure condition (return to its starting state, as a bound eigenstate must) and the propagating condition (never return, as the impedance-match free-photon mode requires). The contradiction is topological and independent of which force implements the closure.
For hydrogen, the gap has a structure that goes beyond the Diophantine proof. The entire photon energy axis \([0, IE)\) divides into four equal quarters of width \(IE/4\), and the distribution of spectral lines across these quarters is the most direct visual confirmation of the two-sector structure of the vacuum.
From the NIST H i catalog (518 observed lines with \(h\nu > 0.1\) meV, \(IE = 13{,}598.435\) meV):
|
Quarter |
Energy range |
Lines |
Series |
|
First |
\([0,\; IE/4) = [0,\; 3{,}399.6)\) meV |
440 |
Balmer, Paschen, Brackett, … |
|
Gap |
\([IE/4,\; 3IE/4) = [3{,}399.6,\; 10{,}198.8)\) meV |
0 |
empty |
|
Fourth |
\([3IE/4,\; IE) = [10{,}198.8,\; 13{,}598.4)\) meV |
76 |
Lyman |
There are zero lines in hydrogen’s middle two quarters. Not one. Not a sparse region. Zero.
The four-quarter partition of H i photon energy space. The energy axis \([0,\,IE)\) divides into four equal quarters of width \(IE/4 = 3{,}399.6\) meV. The first quarter (Balmer and higher series: 440 lines) and the fourth quarter (Lyman series: 76 lines) are populated. The middle two quarters are completely empty. Both occupied bands have identical width \(IE/4\). The gap spans \([IE/4,\,3IE/4)\) with zero lines.
The first quarter is populated by the Balmer series (\(n_f = 2\), all \(n_i \geq 3\)), the Paschen series (\(n_f = 3\)), the Brackett series (\(n_f = 4\)), and all higher series. These are the capacitive-sector transitions: they terminate on shells \(n_f \geq 2\) that support \(\ell \geq 1\) orbital character. The electric mode dominates. The structural primitive \(\varepsilon_0\) organizes the field.
The fourth quarter is populated by the Lyman series only (\(n_f = 1\), all \(n_i \geq 2\)). These are the inductive-sector transitions: they terminate on the \(n = 1\) ground state, which is purely \(\ell = 0\). The magnetic mode organizes the coupling. The non-structural primitive \(\mu_0\) governs the transition.
Between them: two quarters of empty space, spanning \([IE/4, 3IE/4)\), width exactly \(IE/2\).
The empty interval spans exactly \(IE/2\): two full quarters, perfectly centered at \(IE/2\). This symmetry is not accidental. The lower edge at \(IE/4\) is the terminus of the E-sector, organized by \(\varepsilon_0\) (the structural primitive, the source pole \(p\) of the Riemann sphere). The upper edge at \(3IE/4\) is the onset of the H-sector, organized by \(\mu_0\) (the non-structural primitive, the sink pole \(p^*\)). The gap is the spectral interval between the two poles of the Riemann sphere, measured in hydrogen’s own energy units.
The width \(IE/2\) is not a free parameter or a fitted result. It is the consequence of the two-sector structure: the lower edge is fixed at \(IE/4\) (the impedance boundary) and the upper edge is fixed at \(3IE/4\) (the Lyman-\(\alpha\) energy \(= E_0(1 - 1/4) = 3E_0/4\), the minimum energy for a transition terminating on the ground state). Their separation is exactly \(IE/2\).
The Balmer series limit (\(n_i \to \infty\), \(n_f = 2\)) approaches \(IE/4\) from below. The last catalogued Balmer transition (\(n_i = 40\), \(n_f = 2\)) reaches \(3{,}391.1\) meV — still \(8.5\) meV below \(IE/4 = 3{,}399.6\) meV — with oscillator strength proportional to \(40^{-3/2} \approx 4 \times 10^{-3}\) relative to Balmer-\(\alpha\). The transitions closest to crossing the boundary carry the least amplitude. The boundary is approached but never reached, and the approach grows weaker as it does. This is the spectroscopic expression of the impedance mismatch: the field can almost reach the match condition, but the closer it gets, the weaker the transition that would carry it there.
The three lowest distinct states of hydrogen — 1s, 2s, and 2p — encode the complete geometric structure of the impedance boundary in the simplest possible form. They are not an arbitrary selection from the spectrum. The \(1s\) state is the deepest local closure the Coulomb field can sustain: maximum \(Z_{\rm orb}\), maximum impedance mismatch, the source pole \(p\) at its most constrained expression. The \(2s\) state sits at \(\kappa_q\) by construction: its binding energy \(IE/4\) is exactly the impedance boundary, the E/H equilibrium point, the closest any bound state can approach the vacuum match condition. The \(2p\) state is the first orbital with \(\ell \geq 1\) — the first in which the non-structural primitive \(\mu_0\) participates in the closure geometry. These three states therefore span the complete electromagnetic geometry of the vacuum boundary: source pole (\(1s\)), impedance boundary (\(2s\)), first inductive mode (\(2p\)).
Mapping them into binding-energy and angular-momentum coordinates, measured in units of \((IE/4,\, \Delta\ell = 1)\), places the vertices at: \[\text{1s} = (4,\, 0), \quad \text{2s} = (1,\, 0), \quad \text{2p} = (1,\, 1). \label{eq:proto_triangle_vertices}\] This is a right triangle with horizontal leg \(= 3\) (energy: \(4 - 1 = 3\) units of \(IE/4\)), vertical leg \(= 1\) (angular momentum: \(\ell(2p) - \ell(2s) = 1\)), and right angle at \(2s\). The right angle is not a coincidence: it records the orthogonality condition \(\mathbf{E} \perp \mathbf{H}\) at the impedance-match point, the same orthogonality shown in Figure 3.
The horizontal leg has length 3 — an integer. The ideal self-similar ratio from the \(\varphi\)-attractor would give \(\varphi^2 = \varphi + 1 \approx 2.618\), not 3. The departure of the first integer closure from the ideal is: \[3 - \varphi^2 = 3 - (\varphi + 1) = 2 - \varphi = \varphi^{-2}, \label{eq:proto_scar}\] an exact algebraic identity from \(\varphi^2 = \varphi + 1\). The quantity \(\varphi^{-2} \approx 0.382\) is the proto-scar: the geometric fossil of the universe’s first integer closure, visible directly in the energy levels of hydrogen, and related to the same \(\varphi\)-departure that generates \(\alpha\).
The proto-triangle is a geometric consequence, readable directly from the NIST hydrogen spectrum (Kramida et al. 2023). The derivation precedes the triangle; the triangle instantiates it in the simplest possible atom.
The proto-triangle \(\{1s,\,2s,\,2p\}\) in binding-angular space, legs \(3\) and \(1\) in units of \((IE/4,\,\Delta\ell{=}1)\), right angle at \(2s\). The right angle records \(\mathbf{E}\perp \mathbf{H}\) at the impedance-match point (Section 2.6). The deviation \(3 - \varphi^2 = \varphi^{-2}\) (equation [eq:proto_scar], exact) is the proto-scar. The \(2s\) vertex sits at the \(IE/4\) boundary — the impedance-match point no bound-to-bound transition can reach (Section 5.2).
The Diophantine proof holds for hydrogen by arithmetic. The universal claim is that the impedance boundary \(IE/4\) is absent from the transition catalog of every ion in the periodic table — not because of a selection rule peculiar to hydrogen, but because the geometric argument is independent of nuclear charge. The orbital impedance \(Z_{\rm orb} = Z_0\, n/\alpha\) does not depend on \(Z\): the charge and radius cancel from the ratio \(|\mathbf{E}|/|\mathbf{H}|\) exactly as they did in Section 3.4. The impedance boundary is therefore the same condition in every ion, expressed at \(IE/4\) in each ion’s own energy units. An empirical test of 80 ions spanning \(Z = 1\) to \(Z = 82\) confirms the gap in 71 and falsifies it in none (K. B. Heaton and Coherence Research Collaboration 2026c).
What the survey reveals is not merely that the gap exists, but that its width encodes the electromagnetic character of each ion. The gap is wide when the ion’s spectrum is dominated by one sector and narrow when the two sectors are in equilibrium. D-block ions near half-fill of the 3d or 4d subshell — those with \(L = 0\) ground terms, placing them in exact electromagnetic equilibrium at the impedance boundary — carry the narrowest gaps, because the E and H sectors contribute equally and the boundary is most sharply defined. S/p-block ions lie deep in the capacitive sector, far from the boundary, and their gaps are correspondingly wide and diffuse. F-block lanthanides fall between. This ordering — d \(<\) f \(<\) sp — is invisible in absolute electron-volts and becomes visible only on the \(\kappa\) ruler, because it is a property of the vacuum geometry rather than of the individual ions. The vacuum organizes; each ion instantiates.
The gap center \(G = \sqrt{{\rm gap_{lo}} \cdot {\rm gap_{hi}}}\) measures the impedance boundary in each ion’s own energy units. When \(G = IE/4\), the four-constant lock of Section 4.8 gives immediately: \[m_e c^2 = \frac{8G}{\alpha^2}, \qquad a_0 = \frac{\hbar c\,\alpha}{8G}. \label{eq:mass_bohr_recovery}\] These are not fitted formulas. They are algebraic consequences of two things already established: \(G = E_0/4\) (the gap center is one quarter of the Rydberg energy in the infinite-nuclear-mass limit) and \(E_0 = m_e c^2\alpha^2/2\) (the Rydberg energy definition). The electron mass is determined from the location of a structural absence: from where no transition exists, not from where transitions do.
This is the four-constant lock made empirical. Section 4.8 showed that \(\alpha\), \(h\), \(m_e\), and \(a_0\) are algebraically locked — deriving any one from the geometry is equivalent to deriving all four. Equations [eq:mass_bohr_recovery] are the spectroscopic reading of that lock: the gap center \(G\) locates the impedance boundary in energy, and the impedance boundary encodes the mass floor of the orbital hierarchy. The ground state is the deepest closure the Coulomb field can sustain; the gap at \(IE/4\) is where that hierarchy would need to exchange with the vacuum and cannot; and the energy scale of that exchange is \(m_e c^2\alpha^2/8\).
In practice, the mass recovery requires localising the gap boundaries with parts-per-million precision. A denser substrate than observed photon lines is needed: the level-pair analysis uses all pairwise differences of NIST energy levels, giving gap boundaries up to 445 times narrower than the photon-line test from the same data. Five d-block ions — Fe ii, Mo i, Mn ii, Fe i, and Cr ii — satisfy the precision-tier criteria and carry gap widths below 0.1 meV. Their ensemble mean residual, after the proton reduced-mass correction, is (K. B. Heaton and Coherence Research Collaboration 2026c): \[\delta_G^{\rm corr} = -2.8 \pm 2.7~\text{ppm (SEM)},\quad n = 5, \label{eq:headline_result}\] recovering: \[\begin{aligned} \frac{m_e c^2\big|_{\rm gap} - m_e c^2\big|_{\rm CODATA}} {m_e c^2\big|_{\rm CODATA}} &= -2.8 \pm 2.7~\text{ppm}, \label{eq:mass_recovery_result} \\[4pt] \frac{a_0\big|_{\rm gap} - a_0\big|_{\rm CODATA}} {a_0\big|_{\rm CODATA}} &= +2.8 \pm 2.7~\text{ppm}. \label{eq:bohr_recovery_result} \end{aligned}\] These are the energy and length expressions of a single geometric measurement, related by \(m_e c^2 \cdot a_0 = \hbar c/\alpha\). The model makes one claim with four numerical consequences; all four are consistent with measurement.
Why are these particular ions the most precise? Because their \(L = 0\) ground terms place them in exact electromagnetic equilibrium at the impedance boundary. With neither E-sector nor H-sector dominant, the gap straddles \(IE/4\) symmetrically and the geometric mean \(G\) is an unbiased estimator of \(IE/4\). This is not a selection criterion imposed after the fact — it follows from the same two-invariant structure that made the geometric mean the right estimator in Section 4.7.
Gaps are common in atomic spectra. The same test confirms a gap near \(IE/3\) in 78 of 80 ions and near \(IE/2\) in 64 of 80. The existence of a gap at any given fraction is not by itself a physical claim. What is specific to \(IE/4\) is the reason: it is the only fraction at which \(Z_{\rm orb} = Z_0\), and therefore the only fraction at which equations [eq:mass_bohr_recovery] return the correct physical constants with zero free parameters.
The discriminating test is not whether a gap exists but what the gap center encodes. If \(G\) were located at \(IE/3\) rather than \(IE/4\), equation [eq:mass_bohr_recovery] would give \(m_e c^2 = (3/4) \cdot m_e c^2\big|_{\rm CODATA}\), a residual of \(-250{,}000\) ppm. Under identical methodology, the five precision-tier ions measured at \(IE/3\) return \(-250{,}133 \pm 1.1\) ppm and at \(IE/5\) return \(+250{,}129 \pm 4.6\) ppm — in exact agreement with these algebraic predictions (K. B. Heaton and Coherence Research Collaboration 2026c). The pipeline returns \(\pm 250{,}000\) ppm at the wrong fractions and \(-2.8\) ppm at \(IE/4\). The discrimination belongs to the fraction, not to the ions or the method.
The reason is physical, not accidental: \(IE/4\) is where \(Z_{\rm orb} = Z_0\), and no other fraction carries this geometric meaning. The controls confirm that the measurement is reading a real property of the spectrum, not an artifact of the estimator.
The most overdetermined single confirmation is gadolinium, and what it confirms goes beyond the existence of the gap. It confirms that the two sectors — E and H — are geometrically orthogonal in precisely the sense the primitive structure requires.
Gd ii carries the 4f\(^7\) half-filled configuration, which gives \(L = 0\): complete orbital cancellation, neither E-sector nor H-sector dominant, the gap straddling \(IE/4\) symmetrically. Three completely independent instruments locate the same boundary: NIST optical spectroscopy at \(10^{14}\) Hz, electron paramagnetic resonance at \(10^{10}\) Hz, and the E/H balance method from the level structure alone (K. B. Heaton and Coherence Research Collaboration 2026c). The three instruments span six orders of magnitude in frequency and share no experimental method. They agree within 0.014 \(\kappa\)-units of \(\kappa_q\). The boundary is real and it is the same boundary.
The EPR result carries a specific geometric interpretation. Gd\(^{3+}\) has \(g = 2.000\) and pure spin relaxation character, with no orbital angular momentum contribution. In the language of this paper: the ion couples to external magnetic fields exclusively through the non-structural primitive \(\mu_0\) (the spin mode), with zero contribution from the structural primitive \(\varepsilon_0\) (the orbital mode). This is the \(L = 0\) condition expressed as a coupling selection. The impedance boundary appears at EPR frequencies because the boundary is a property of the vacuum geometry, not of the optical regime.
The complementary case is Ho ii (4f\(^{11}\), \(L = 6\)), whose spectral mass lies almost entirely in the inductive H-sector. An E-sector instrument — the Coulomb-closure template that searches for capacitive transitions — should return zero resonant pairs for this ion, not because the NIST catalog is sparse, but because the ion is electromagnetically orthogonal to the probe. The prediction is confirmed: Ho ii returns zero hitpairs at every \(\kappa\)-bin despite carrying more NIST energy levels than Mn ii, which returns 63,134 hitpairs at the same resolution (K. B. Heaton and Coherence Research Collaboration 2026c).
The null result is not an absence of data. It is the predicted outcome of geometric orthogonality: the E and H modes of the vacuum are perpendicular, an E-sector instrument is blind to H-sector organization, and Ho ii is organized entirely by the H-sector. This is the same orthogonality shown in Figure 3 for the free vacuum, now confirmed in the spectroscopic record of a specific ion.
What Section 5 has established.
Every section of this paper so far has operated within a silent assumption. In the Thévenin circuit of Section 3, the bottom wire — the H-mode return path, the non-structural primitive \(\mu_0\) closing the loop — is drawn as a short circuit. Voltage zero. No impedance. The electron removed to the continuum falls to nothing.
This is not true. The electron removed to the continuum falls to \(m_e c^2\).
The mass of the electron is not zero. It is not a small correction. It is the minimum energy that any local closure of the electromagnetic field can carry — the geometric floor below which no stable configuration exists, below which the closure geometry itself dissolves. Everything above this floor is the spectroscopy we have been deriving. The floor itself is the mass.
This section establishes the relationship between impedance, geometry, and mass as a principle. Not a coincidence of units, not a dimensional argument, but a structural consequence of the same geometry that generates the gap at \(IE/4\). The mass equation \(m_e c^2 = 8G/\alpha^2\) is already encoded in what we have derived. We now make it explicit, show where it comes from geometrically, and extend the picture to the relativistic boundary and to the nuclear scale.
The Thévenin circuit of Figure 4 is exact within its domain. That domain is the bound spectrum: the set of all closure modes at quantum numbers \(n \geq 1\), confined below the ionization energy \(IE\). Within this domain, the circuit correctly predicts the impedance mismatch, the reflection coefficient, the gap at \(IE/4\), and the power available for photon emission. It is complete for the questions it can ask.
But there is a question it cannot ask: what happens when the electron crosses the ionization boundary and enters the continuum?
In the circuit picture, ionization corresponds to the load \(Z_0\) becoming the entire universe — the electron is no longer in a bound closure, it is in the free propagating field. The source voltage \(V_{\rm TH} = IE\) has been fully supplied, and the current flows into the vacuum load without returning. The bottom wire — the return path — should carry the energy that the electron retains after ionization.
In the non-relativistic model, we set this to zero. The ionised electron has “escaped” and we assign it zero residual energy in the circuit picture. This is the short-circuit approximation of the bottom wire: the return path has no impedance because we are not tracking what happens after ionization.
But the electron does not have zero energy after ionization. It has \(m_e c^2\) at minimum: the rest energy that it carries regardless of its kinetic state, regardless of its distance from the nucleus, regardless of any external field. The true ground of the circuit is not voltage zero. It is the mass floor.
When we draw the bottom wire as a short, we are implicitly setting \(m_e c^2 = 0\). This is the non-relativistic approximation, and it is exact for all questions confined to the bound spectrum (\(\Delta E \ll m_e c^2\)). It fails precisely when we ask what the mass is — because the mass is what the bottom wire carries, and we have set it to zero.
The deepest closure mode the Coulomb field can sustain is \(n = 1\): the ground state, with orbital impedance \(Z_{\rm orb}(1) = Z_0/\alpha \approx 137\, Z_0\). This is the maximum impedance mismatch, the minimum transmission coefficient, the most tightly bound local closure.
Can the field sustain a closure at \(n < 1\)? The formula \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) gives \(Z_{\rm orb} < Z_0/\alpha\) for \(n < 1\). But the principal quantum number must be a positive integer. There is no \(n = 0\). There is no closure mode between the ground state and the continuum.
This is the geometric floor. Below the ground state, the closure geometry dissolves: there is no standing wave that the Coulomb field can sustain against the vacuum. The field configuration that would correspond to “\(n = 0\)” is not a more tightly bound orbital. It is the free propagating mode itself — \(Z_{\rm orb} = Z_0\), the photon. The integer constraint \(n \geq 1\) is what keeps the bound configuration from collapsing all the way to the propagating mode.
The energy of the ground state closure is \(IE\) (the ionization energy of the atom). The energy of the configuration just below the ground state — the energy of the transition from a bound closure to no closure at all — is \(m_e c^2\): the rest energy that the electron carries when it is no longer in any closure mode, when it is simply a free field configuration with its intrinsic mass.
The relationship between the ground state energy \(IE\) and the mass floor \(m_e c^2\) is therefore set by the impedance hierarchy: \[\frac{IE}{m_e c^2} \sim \alpha^2, \label{eq:IE_mass_ratio}\] which follows from \(IE = E_0 = m_e c^2 \alpha^2/2\) (the Rydberg energy relation). The ionization energy is suppressed below the mass floor by two factors of \(\alpha\): one for the orbital velocity \(v_n/c = \alpha/n\), and one for the impedance ratio \(Z_0/Z_{\rm orb} = \alpha/n\). The bound state is a doubly \(\alpha\)-suppressed version of the mass floor geometry.
The mass floor is not external to the impedance hierarchy. It is the hierarchy’s own ground: the energy scale at which the local closure geometry ceases to exist and the field returns to the free-propagating configuration. The electron rest mass \(m_e c^2\) is the energy of the electromagnetic field at that configuration.
We have already derived the relationship between the gap center \(G\) and the electron rest mass. It is worth pausing to see exactly where it comes from geometrically, because the derivation is simpler than it appears.
The gap center \(G = \sqrt{{\rm gap_{lo}} \cdot {\rm gap_{hi}}}\) measures the geometric mean of the two gap boundaries in the \(\kappa\) coordinate. For a symmetric ion in the infinite-nuclear-mass limit, \(G = E_0/4\): the gap center equals one quarter of the Rydberg energy.
The Rydberg energy is defined by: \[E_0 = \frac{m_e c^2\, \alpha^2}{2}. \label{eq:rydberg_definition}\] Substituting \(G = E_0/4\): \[G = \frac{m_e c^2\, \alpha^2}{8} \quad \Longrightarrow \quad \boxed{m_e c^2 = \frac{8G}{\alpha^2}.} \label{eq:mass_from_gap}\] And from \(E_0 = \hbar c\,\alpha / 2a_0\) (the Rydberg energy in terms of the Bohr radius): \[G = \frac{E_0}{4} = \frac{\hbar c\,\alpha}{8 a_0} \quad \Longrightarrow \quad \boxed{a_0 = \frac{\hbar c\,\alpha}{8G}.} \label{eq:bohr_from_gap}\]
These two equations express the same geometric fact: the gap center \(G\) is the energy scale at which the orbital impedance hierarchy intersects the vacuum impedance boundary. That intersection energy, times \(8/\alpha^2\), is the mass floor. The Bohr radius is the length scale corresponding to that same intersection. These two equations demonstrate that the electron rest mass and the Bohr radius are encoded in the impedance geometry as consistency conditions: a gap center correctly located at \(IE/4\) returns both constants simultaneously, to parts-per-million precision, with no free parameters. The circuit has a floor, the floor has an energy, and that energy is \(m_e c^2\) — not as an input assumed in advance, but as the value the geometry requires to be self-consistent. Whether this consistency can be sharpened into a full first-principles derivation of \(m_e\) from the standing-wave structure of the orbital field at the gap boundary and without reference to any measured atomic constant is the natural next question, and the one the companion work in preparation addresses.
The gap center \(G\) is more than a numerical midpoint between two catalog entries. It is the geometric mean of two sector boundaries — the capacitive terminus \({\rm gap_{lo}}\) (last E-sector transition below \(IE/4\)) and the inductive onset \({\rm gap_{hi}}\) (first H-sector transition above \(IE/4\)). As established in Section 4.7, \(G\) is structurally parallel to \(Z_0 = \sqrt{\mu_0/\varepsilon_0}\): both are geometric means of two complementary sector constants.
At a deeper level, \(G\) is the spectroscopic expression of the orbital field as a standing wave governed by Maxwell’s equations at the impedance boundary. The levels whose differences bracket \(IE/4\) are not arbitrary energy eigenvalues. They are the standing wave configurations of the orbital field at the frequencies closest to the impedance match condition \(Z_{\rm orb} = Z_0\). The geometric mean of their energy differences is the field-geometric balance point — the configuration at which the E and H sectors of the orbital field are in equilibrium with each other and with the vacuum.
The full characterization of \(G\) as a standing wave impedance geometry — including the internal E-H balance of the orbital field and its coupling to the external vacuum — is the next layer of the theory. The present paper establishes the principle: \(G\) is the impedance boundary expressed in each ion’s own energy units, and the mass and Bohr radius are determined by where that boundary sits. The precise field-geometric derivation of \(G\), including the role of angular momentum in displacing the gap center from \(IE/4\) in electromagnetically asymmetric ions, is the subject of a companion paper in preparation.7
We can now state the four-constant lock in its final form. The four fundamental constants of atomic physics — \(\alpha\), \(h\), \(m_e\), and \(a_0\) — are not independent measurements. They are four expressions of a single geometric fact: the scale at which the self-similar logarithmic field on the Riemann sphere encounters the constraint of integer closure.
\[\begin{aligned} \alpha &= \frac{Z_0}{2R_K} && \text{the departure from } \varphi\text{-attractor} \nonumber \\ h &= \frac{E(\kappa)}{\nu(\kappa)} && \text{the scaling invariant of the self-similar field} \nonumber \\ m_e c^2 &= \frac{8G}{\alpha^2} && \text{the geometric floor of the impedance hierarchy} \nonumber \\ a_0 &= \frac{\hbar c\,\alpha}{8G} && \text{the length scale of that floor} \nonumber \end{aligned}\]
Each is derived from the geometry. None is a free parameter. Their mutual consistency — confirmed to \(-2.8 \pm 2.7\) ppm (SEM) across five precision-tier ions — is not four separate numerical agreements. It is one geometric structure, read four times by four different instruments.
The relationship between the non-relativistic orbital model and the full relativistic picture is most clearly shown as a sequence of three circuit diagrams, each adding one layer of geometric structure.
The three-circuit progression from non-relativistic orbital to relativistic completion. Circuit 1 (non-relativistic orbital): Thévenin source \(V_{\rm TH} = IE\), series impedance \(Z_{\rm orb} = Z_0\,n/\alpha\), load \(Z_0\), bottom wire a short circuit. Exact for all bound-spectrum transitions where \(\Delta E \ll m_e c^2\). The mass floor is present in principle but suppressed to zero in the limit \(IE/m_e c^2 = \alpha^2/2 \to 0\). Circuit 2 (ionization boundary): the bottom wire acquires impedance \(Z_m = Z_0\cdot 2/\alpha^2 \approx 14{,}149{,}000\;\Omega\), the geometric floor of the impedance hierarchy. The electron crossing the ionization boundary carries minimum rest energy \(m_e c^2\); the gap at \(IE/4\) encodes this floor through \(m_e c^2 = 8G/\alpha^2\). Circuit 3 (relativistic completion): the source voltage becomes the frame-dependent total energy \(E_{\rm tot} = \sqrt{(m_e c^2)^2 + (pc)^2}\). The top-rail impedance \(Z_{\rm orb}\) is unchanged. The bottom-wire impedance \(Z_m\) (dashed red border) is identical to Circuit 2 in value — but is now identified as the Lorentz scalar: the fiber-constant element that every Lorentz boost leaves untouched, the invariant norm \(m_e c^2 = \sqrt{E_{\rm tot}^2 - (pc)^2}\) of the four-momentum. The Minkowski triangle (lower right) shows the same identity geometrically: \(m_e c^2\) is the red vertical leg that persists unchanged as \(pc\) grows from zero (rest frame, Circuit 2) to any relativistic value; \(E_{\rm tot}\) and \(pc\) are the frame-dependent sides. The dashed red border on \(Z_m\) and the red leg of the triangle are the same object in two vocabularies. The three circuits trace one geometric statement: mass is not an input to the electromagnetic geometry. It is the floor of the impedance hierarchy (Circuit 2), and the fiber-constant norm that every Lorentz boost leaves untouched (Circuit 3). The bottom wire is the Lorentz scalar. The field equation that the spectrometer and circuit instruments read from this impedance structure is identified in Figure 5.
The Thévenin source \(V_{\rm TH} = IE\), series impedance \(Z_{\rm orb} = Z_0\, n/\alpha\), load \(Z_0\), bottom wire = short. This is the circuit of Sections 3 through 5. It is exact for all transitions within the bound spectrum, where \(\Delta E \ll m_e c^2\). The mass floor is present but suppressed: setting the bottom wire to zero is equivalent to working in the limit \(\alpha^2 \to 0\), where \(IE/m_e c^2 = \alpha^2/2 \to 0\) and the mass floor lies infinitely far below the bound spectrum in natural units.
When the electron crosses the ionization boundary, the bottom wire acquires a non-trivial geometric depth. The electron entering the continuum carries energy \(m_e c^2\) at minimum, regardless of its kinetic state. This rest energy defines the floor of the impedance hierarchy: the depth below which no closure geometry can be sustained. It is expressed as the dimensionless ratio of the mass energy to the ionization energy: \[\frac{m_e c^2}{IE} = \frac{m_e c^2}{E_0} = \frac{2}{\alpha^2} \approx 37{,}558. \label{eq:mass_depth}\] This ratio is not a physical impedance in ohms. It is the depth of the mass floor in the \(\alpha\)-hierarchy: the number of \(Z_0\)-units that separate the free-propagating field from the mass floor, just as \(Z_{\rm orb}(1)/Z_0 = 1/\alpha \approx 137\) is the depth of the ground-state orbital above the free field. The mass floor sits \(2/\alpha \approx 274\) times deeper than the orbital impedance mismatch at \(n = 1\). In the full hierarchy: \[\begin{aligned} \text{Free field:} &\quad Z_0 \approx 377\,\Omega \nonumber \\ \text{Ground state orbital:} &\quad Z_{\rm orb}(1) = \frac{Z_0}{\alpha} \approx 51{,}626\,\Omega \nonumber \\ \text{Mass floor:} &\quad Z_0 \cdot \frac{2}{\alpha^2} \approx 14{,}149{,}000\,\Omega. \label{eq:hierarchy} \end{aligned}\] Each level is suppressed by one additional factor of \(\alpha\) relative to the one above it. The mass floor is a doubly \(\alpha\)-suppressed boundary: once for the orbital velocity (\(v/c = \alpha\)) and once for the impedance ratio (\(Z_0/Z_{\rm orb} = \alpha\)). It is the ultimate limit — no electromagnetic closure can sustain itself below this depth, because below it the geometry that makes closure possible has dissolved.
The source voltage becomes \(E_{\rm tot} = \sqrt{(m_e c^2)^2 + (pc)^2}\): the frame-dependent total energy of the ionised electron in a frame where it carries momentum \(p\). The top-rail impedance \(Z_{\rm orb}\) is unchanged — the orbital geometry is the same. What changes is only the identity of the bottom wire.
In Circuit 2, \(Z_m\) was the mass floor of the impedance hierarchy, expressed as a ratio. In Circuit 3, \(Z_m\) is recognized as something more fundamental: it is the Lorentz scalar of the energy-momentum system. The Lorentz boost is a rotation in the \((E_{\rm tot},\, pc)\) plane — it redistributes energy between the two frame-dependent quantities while leaving their difference \(E_{\rm tot}^2 - (pc)^2 = (m_e c^2)^2\) invariant. In the language of the companion paper on determinacy under quotient representations (K. B. Heaton and Coherence Research Collaboration 2026b), compressing \((E_{\rm tot},\, pc)\) to velocity \(v = pc/E_{\rm tot}\) discards \(m_e c^2\), which lives in the fiber of the Lorentz representation. \(Z_m\) is the circuit element that is not discarded.
The Minkowski triangle in Figure 8 shows this geometrically. The red vertical leg \(m_e c^2\) is the same length in every Lorentz frame: as \(pc\) grows from zero (rest frame, Circuit 2) to any relativistic value, the hypotenuse \(E_{\rm tot}\) grows with it, but the vertical leg is untouched. The dashed red border on \(Z_m\) in the circuit diagram and the red leg of the triangle are the same object in two vocabularies. The circuit identifies the physical element; the triangle shows its invariance.
The gap at \(IE/4\) is therefore the spectroscopic approach to this floor. The transition that would carry energy \(IE/4\) is the transition that comes closest to bridging the gap between the bound closure geometry and the free-propagating geometry. That it is forbidden by number theory (no integer \(n\) satisfies the closure condition at that energy) is the spectroscopic record of the mass floor’s location in the hierarchy — and that floor is, in Circuit 3’s terms, the Lorentz scalar the bound spectrum reads from above.
The full relativistic treatment — in which the invariant mass labels the persistent energy-momentum constraint surface and the bridge between \(Z_m\) as a circuit impedance and \(m_e c^2\) as a relativistic invariant is made derivable rather than merely stated — is developed in (K. B. Heaton and Coherence Research Collaboration 2026b). What the present paper establishes is the principle and the picture: mass is not an external input to the electromagnetic geometry. It is the geometry’s own floor, the fiber-constant element of the Lorentz representation, readable from the gap at \(IE/4\) through \(m_e c^2 = 8G/\alpha^2\), and visible in Circuit 3 as the one element of the orbital circuit that no change of frame can alter.
The atomic gap at \(IE/4\) could, in principle, be a peculiarity of the electron-Coulomb system: a consequence of the specific force law and mass scale rather than of the vacuum geometry. The nuclear scale provides the decisive test. If the \(1/4\) fraction belongs to the vacuum geometry rather than to atomic physics, the same boundary should appear wherever a closure condition is satisfied at \(\kappa_q\) — regardless of which force implements the closure and regardless of the mass scale involved.
It does.
The nuclear shell model (Goeppert Mayer 1949; Haxel, Jensen, and Suess 1949) established that doubly magic nuclei — those with both proton and neutron numbers at simultaneous shell closure — exhibit anomalously large gaps above the ground state. In the language of this paper, these are configurations where the nuclear closure condition is satisfied at the impedance boundary: the simultaneous closure of both shell sequences removes all low-lying excitation channels below a specific energy threshold. That threshold is \(S_n/4\), where \(S_n\) is the neutron separation energy — the nuclear analog of \(IE\).
The \(1/4\) fraction is not imposed. It follows from the same Diophantine structure that governs the atomic case. The mathematical argument is independent of the force that implements the closure: what matters is that the closure condition is satisfied at integer quantum numbers, and that the same impossibility theorem applies. The vacuum geometry organizes the threshold; each scale’s force law determines whether a given configuration satisfies the closure condition at that threshold.
All four doubly magic nuclei surveyed (\(^{16}\)O, \(^{48}\)Ca, \(^{132}\)Sn, \(^{208}\)Pb) show complete depletion of observed levels below \(S_n/4\), with first excited states well above the boundary (K. B. Heaton and Coherence Research Collaboration 2026c). Of 20 non-doubly-magic nuclides surveyed, none shows a gap. The two populations are cleanly separated with zero overlap: closure satisfied gives a gap; closure not satisfied gives no gap. The boundary is at \(\kappa_{\rm nuc} \approx 0.180\) for the doubly magic population and resumes immediately at \(\kappa_q \approx 0.282\) for the mid-shell population — a separation of \(\Delta\kappa = 0.100\) with no intermediate cases.
The sharpest single illustration of what the nuclear argument proves is gadolinium. Atomic Gd ii carries one of the sharpest atomic gaps in the survey, because its 4f\(^7\) half-filled configuration achieves \(L = 0\): the atomic closure condition is satisfied at the impedance boundary. Nuclear gadolinium (\(Z = 64\), mid-shell between magic 50 and 82; \(N = 91\)–\(94\), mid-shell between magic 82 and 126) carries no nuclear gap: the nuclear closure condition is not satisfied at \(S_n/4\) (K. B. Heaton and Coherence Research Collaboration 2026c).
Same element. Same \(\kappa_q\). Two independent closure conditions, one satisfied and one not. Two opposite results at the same boundary.
This decoupling proves the principle precisely: the \(1/4\) boundary belongs to the vacuum geometry, not to any particular force or configuration. It appears where closure is achieved and is absent where closure fails. The force that implements the closure — electromagnetic at the atomic scale, strong at the nuclear scale — is irrelevant to the geometry of the boundary.
The nuclear argument completes the mass principle. At the atomic scale, the geometric floor is \(m_e c^2\), encoded in the gap through \(m_e c^2 = 8G/\alpha^2\). At the nuclear scale, the analogous floor is the nuclear binding energy per nucleon, encoded through the same \(\kappa_q\) applied to \(S_n\). These are not two separate facts. They are two readings of the same geometric principle at two different scales: mass is the energy of the field configuration that sits at the closure floor, and the closure floor sits at \(\kappa_q\) whenever the relevant closure condition is satisfied.
The vacuum geometry organizes both scales. The mass floors at each scale are its signatures. What changes between the atomic and nuclear cases is not the geometry — \(\kappa_q\) is the same — but the force that achieves closure and the mass scale that results. The principle is universal because the geometry is universal: it belongs to the vacuum, not to the matter that instantiates it.
The derivation of \(Z_{\rm orb}(n) = Z_0\,n/\alpha\) in Section 3.4 contained an algebraic observation whose deeper meaning was deferred. When the Coulomb electric field and the Biot–Savart magnetic field were formed into the ratio \(|\mathbf{E}|/|\mathbf{H}|\), both the elementary charge \(e\) and the orbital radius \(r_n\) cancelled exactly from the numerator and denominator. The impedance ratio depends only on \(n\) and \(\alpha\). This is worth pausing on. The orbital carries charge; the orbital has a radius; yet neither enters the quantity that determines how bound the orbital is relative to the vacuum. What remains is purely geometric: the winding count \(n\) and the departure from the \(\varphi\)-attractor, \(\alpha\).
This cancellation is not a convenience. It is the spectroscopic record of what charge is in the framework established in Section 1. Theorem 37 of (Coherence Research Collaboration 2026) assembles \(\alpha = Z_0/2R_K\) from four facts; the third of those facts establishes \(e\) as the minimum topological charge of the defect at the source pole \(p\), via the Dirac quantisation condition on the punctured sphere. One complete circuit around \(p\) carries exactly one quantum of charge, which is the same residue \(\mathrm{Res}(1/z, 0) = 1\) that Section 1 identified as the irreducible structural unit persisting at the singularity. The charge is what the winding produces. In this framework, \(e\) is not an independent primitive of the electromagnetic vacuum; it is determined by \(\alpha\), \(Z_0\), and \(R_K\) through an exact identity (Coherence Research Collaboration 2026). Charge is not a free parameter of the impedance geometry. It is the topological count of how many times the closure winds around the source pole: always one, for every orbital at every \(n\).
The fact that \(e\) cancels from \(Z_{\rm orb}\) is therefore the algebraic expression of a structural fact: charge is topological, and the impedance is geometric. The ratio \(|\mathbf{E}|/|\mathbf{H}|\) measures the geometric mismatch of the bound field relative to the vacuum; it is blind to the topological count of the closure, just as the Riemann sphere’s logarithmic field is blind to the number of times an orbit winds around \(p\) once the winding is closed. Two orbitals differing in \(n\) differ in their geometric mismatch by the factor \(n\). They carry the same topological charge because both are single-winding closures of the same primitive defect.
With charge identified as topological, the staircase structure of the orbital hierarchy becomes visible in a new way. The hydrogen self-resonance portrait discussed in Section 7 shows dominant collision rungs at \(\kappa(n) = 2\log_{10}n\,/\,|\log_{10}\alpha|\) for \(n = 1, 2, 3, \ldots\): the ionization limit at \(\kappa = 0\), the impedance boundary at \(\kappa_q \approx 0.282\), the third-shell limit at \(\kappa \approx 0.448\), and so on. The staircase descends from higher \(\kappa\) toward \(\kappa = 0\) as \(n\) decreases, reaching the ground state at \(n = 1\). There it stops. Not because there is nothing below — but because \(n = 1\) is the minimum positive integer, and the orbital hierarchy is indexed by positive integers. The staircase terminates at \(\kappa = 0\) by the same integer necessity that creates the gap at \(\kappa_q\): the closure condition requires integer winding, and there is no integer below \(1\).
What sits below \(\kappa = 0\) is not an absence. It is the platform the staircase rests on.
The mass floor derived in Section 6.2 locates this platform precisely: \[\kappa_{\rm mass} \;=\; -\!\left(2 + \frac{\log_{10}2}{|\log_{10}\alpha|}\right) \;\approx\; -2.141. \label{eq:kappa_mass}\] The structure of this expression repays attention. The \(-2\) is exact: it reflects two factors of \(\alpha\) separating the orbital hierarchy from the mass floor, one from the closure velocity (\(v_n/c = \alpha/n\), evaluated at \(n = 1\)) and one from the impedance ratio (\(Z_0/Z_{\rm orb}(1) = \alpha\)). This is the doubly \(\alpha\)-suppressed boundary that Section 6.2 named without yet explaining its structure. The \(-0.141\) is the contribution of the factor of \(2\) in the Rydberg definition \(E_0 = m_e c^2\alpha^2/2\): had the Rydberg energy been \(m_e c^2\alpha^2\) rather than \(m_e c^2\alpha^2/2\), the mass floor would have fallen at exactly \(\kappa = -2\). In either case, the result is the same in physical content: the mass floor is two complete \(\alpha\)-steps below the ionization limit, separated from the orbital hierarchy by the same factor \(\alpha^2/2 = IE/m_e c^2\) that the Rydberg energy relation has always carried.
The electron is the closure whose floor the orbital hierarchy is reading from above. A closure at energy \(IE_{\rm electron} = m_e c^2\) would have its own \(\kappa_q\) boundary at \[\frac{m_e c^2}{4} \;=\; 127.75\;\mathrm{keV},\] by the same argument that places the atomic \(\kappa_q\) at \(IE/4\) (Section 5). The force that implements the electron’s self-closure is not electromagnetism at the Bohr scale; it is whatever physical process sustains the electron as a stable configuration at the Compton scale. The geometry, however, is the same. The \(1/4\) fraction belongs to the vacuum, not to the force.
The equation \(m_e c^2 = 8G/\alpha^2\) of Section 6.3 is then not only a recovery of a constant from a spectroscopic gap. It is the orbital hierarchy reading the electron’s closure floor from above through the impedance boundary that the two hierarchies share. The gap at \(IE/4\) is the atomic hierarchy’s closest approach to the electron’s own \(\kappa_q\) boundary: the energy at which the atomic closure mode would momentarily become the electron closure mode, if any integer could satisfy the closure condition there. None can. The gap is the spectroscopic record of that impossibility — and the electron mass is what the gap encodes, precisely because the electron is what occupies the floor below.
There is an important limitation of the coordinate system to name here, because it bears directly on what can and cannot be tested with the tools developed in this paper. The Thread Frame \(\kappa\) coordinate of Section 4 is built on Rydberg energy spacings — E-sector transitions organized by the structural primitive \(\varepsilon_0\). The hydrogen self-resonance portrait is therefore the E-sector face of the orbital hierarchy. The H-sector face — the Lyman series, organized by \(\mu_0\) and terminating on the \(\ell = 0\) ground state — is not accessible to the Thread Frame, for the same reason that Ho ii returned zero hitpairs in the E-sector sweep (Section 5.8): the two sectors are orthogonal, and the E-sector instrument is blind to H-sector organization. What sits below \(\kappa = 0\) in the electron’s own hierarchy requires a probe built on the electron’s Compton-scale structure rather than the Bohr-scale Coulomb hierarchy. Naming this limitation is not a qualification; it is the framework specifying exactly what the next instrument needs to be.
Three independent scales now confirm the universality of the \(1/4\) fraction: atomic (\(IE/4\), a few electron-volts), electron (\(m_e c^2/4 = 127.75\) keV, between atomic and nuclear energies), and nuclear (\(S_n/4 \approx 2\) MeV, Section 6.5). All three share the same \(\kappa_q = 0.282\) because the fraction belongs to the vacuum geometry. The proton’s \(\kappa_q\) sits at \(m_p c^2/4 \approx 235\) MeV, in the range of hadronic excitation spectroscopy. Deriving the proton-to-electron mass ratio \(m_p/m_e \approx 1836\) from this framework requires the strong-force closure geometry, which is non-perturbative in a way that electromagnetic closure is not. What the framework establishes is the principle and the hierarchy: mass is the geometric floor of a closure implemented by a force, \(\kappa_q = 0.282\) is the universal signature of that floor at every scale where the closure condition is satisfied, and the mass ratio is the ratio of two closure-scale energies whose individual values are now named precisely enough to be looked for. That is not a mystery in waiting. It is a program.
We have now traced the complete chain: from two irreducible primitives and their encounter on the Riemann sphere, through Maxwell’s equations as the interior description of that encounter, through the orbital impedance and the Thévenin circuit, through the photon as the calved \(Z_0\) relationship, through the \(\kappa\) coordinate and the Harmonic Light Law \(\nabla^2(\log\nu) = 0\), and the gap at \(IE/4\), to the recovery of the electron rest mass and Bohr radius from the location of a structural absence in 71 atomic spectra. The chain has no free parameters, no fitted constants, and no falsifications.
This section steps back from the derivation and asks five questions that the derivation itself cannot answer:
Modern physics is spectacularly successful. Maxwell’s equations and quantum mechanics both work, and so does the Standard Model. The fine-structure constant has been measured to eleven significant figures, the electron mass is known to parts per billion, and the Bohr radius is calculated from first principles to precisions that strain the limits of measurement. The remaining mysteries — why \(\alpha \approx 1/137\), why the vacuum has exactly two constitutive constants, why the periodic table has the structure it does — were not failures of physics. They were the next question. The answer had been present in Maxwell’s equations since 1865, inside a representational choice made so early and practiced so universally that it had ceased to look like a choice: the compression of two independent vacuum constants into one.
When Johannes Kepler stopped insisting that planetary orbits were circles and allowed the geometry to be what it actually was, the laws of planetary motion followed necessarily. The data was already complete — no new planets, new forces, or new measurements were required. What changed was Kepler’s willingness to read the geometry without imposing a prior assumption onto it. The mathematics and the observations had always been telling the truth. The question was whether anyone was prepared to hear it.
The present work followed a similar path. The two independent constants of the electromagnetic vacuum were always present in Maxwell’s equations, but the compression of \((\varepsilon_0, \mu_0)\) into \(c\) alone was made so early and practiced so universally that it ceased to look like a choice at all. When both constants are restored and the geometry they generate is read without remainder, the following cease to be mysterious inputs and become necessary consequences: the value of the fine-structure constant, the electron rest mass, the Bohr radius, Planck’s constant, and the organization of every atomic and nuclear spectrum in the periodic table.
The result is a geometric unification at the electromagnetic scale — not a dynamical unification of forces, but the identification of a single underlying structure from which all of these follow with zero free parameters, zero new physical entities, and zero falsifications. The unification does not merge forces or introduce new symmetries. It identifies the geometry that was always implicit in the equations and shows what follows from it when both of its independent constants are retained.
What makes this a genuine discovery rather than a reinterpretation is the empirical content: 71 atomic ions confirmed, zero falsified, the electron rest mass recovered to parts-per-million from the location of a structural absence at \(IE/4\), and nuclear corroboration at a second independent physical scale. The geometry makes predictions. The predictions hold. A framework that derives more from fewer assumptions is strictly stronger than one that requires more to derive less, and the predictions of this framework were not fitted to the data. They were calculated from the geometry alone and then checked.
Thus the fine-structure constant is no longer a mysterious number. It is the departure of the vacuum’s winding ratio from the golden-ratio attractor, forced by the irremovable defect in the structural primitive. The electron mass is not an independent measured input. It is the geometric floor of the impedance hierarchy, encoded in every atomic spectrum at the location \(IE/4\). The Bohr radius is not a separate physical scale. It is the length expression of the same floor. Planck’s constant is not a quantum postulate. It is the scaling invariant of the self-similar logarithmic field — the ratio that must be the same at every scale when energy and frequency both scale as \(\alpha^\kappa\).
Why two primitives began to relate through a flaw is a question we will leave a mystery.
The Harmonic Light Law \(\nabla^2(\log\nu) = 0\) is not a law imposed on the vacuum from outside. It is what the vacuum cannot help but be. The logarithmic field between the two poles of the Riemann sphere satisfies Laplace’s equation by construction; energy couples to that field exponentially; Planck’s relation connects energy to frequency; and therefore the logarithm of photon frequency inherits the same harmonic character as the field itself. The law is inevitable because the geometry is inevitable.
Throughout this paper, we have identified \(\alpha\) in three roles:
|
Domain |
Quantity |
\(\alpha\) appears as |
|
Riemann sphere |
\(\xi_{\rm actual} = \varphi - \alpha\) |
departure from \(\varphi\) |
|
Spectral / logarithmic |
\(\beta = \log_{10}\alpha\) |
slope of frequency backbone |
|
Circuit / admittance |
\(-1/Z_{\rm orb} = -\alpha/(Z_0 n)\) |
slope of I–V characteristic |
These are connected by the exact identity: \[-\frac{1}{Z_{\rm orb}(n)} = \frac{10^\beta}{Z_0\, n}, \label{eq:beta_zorb_identity_disc}\] since \(10^\beta = 10^{\log_{10}\alpha} = \alpha\). The spectrometer and the circuit meter are reading the same quantity — and the reason is structural. The Thread Law is the unique solution to \(\nabla^2(\log\nu) = 0\) with a source at \(p\): the only slope consistent with the field’s boundary conditions on the sphere. The I–V characteristic is the same equation read by the circuit: \(V = E_0\,e^{-\phi}\) and \(dI/dV = -1/Z_{\rm orb}\). Both instruments are aimed at the same geometry.
The empirical reach of this geometry is broader than the atomic case alone. An independent information-theoretic analysis (Kelly B. Heaton and Coherence Research Collaboration 2025a) asked purely: which slope organizes photon frequency data most compressively, with no reference to the Thread Frame or the \(\kappa\) coordinate? Across six independent spectral domains — laboratory atomic spectra (Ne i, Na i, Hg ii), the solar spectrum (NSO/Kitt Peak FTS), the stellar spectrum of Vega (ELODIE archive), and molecular bands of C\(_2\) (ExoMol) — the answer was \(\log_{10}\alpha\). The Thread Law \(\nu(\kappa) \propto \alpha^\kappa\) is not a consequence of choosing convenient units. It is a relationship the data obeys regardless of how it is organized, confirmed across atomic, solar, stellar, and molecular domains with zero free parameters. The study confirms that photon frequencies in nature are organized as a harmonic function — which is precisely what \(\nabla^2(\log\nu) = 0\) requires them to be, for any photon that is a \(Z_0\) relationship propagating in the vacuum geometry.
There is a deeper way to state what the Harmonic Light Law means, one that the formalism makes precise but does not make vivid. The vacuum field is an analog substrate: a continuous, smooth, self-similar harmonic potential with no preferred scale, no smallest unit, no floor below which structure disappears. It does not know about integers. It does not know about quanta. It is the pure relationship between two primitives, propagating at \(c\), and that relationship is a continuum all the way down.
Within the chain derived in this paper, the discrete enters at two specific moments, both as boundary conditions imposed on the continuous field rather than as properties of the field itself. First, when the continuous field attempts to close on itself: closure requires an integer winding number, and the integers are the arithmetic of what “return to starting point” means for a wave. The discrete energy levels of the atom are born here — not as a property of the field, but as the scar left by the closure condition on an otherwise continuous geometry. Second, when a piece of the field crosses the orbital boundary and reorganizes into the only configuration the vacuum supports: the photon. The photon is not the fundamental object. It is the minimum transferable unit of the continuous field across a boundary — a sample taken from the analog substrate by the act of closure.
This is why no matter how finely any instrument probes the vacuum, it finds wave-like structure beneath. The waves are not a feature of matter. They are the prior condition. Matter — the atom, the electron, the photon — is what the continuous harmonic field looks like when it is forced to declare a preferred scale by the topology of closure. The quanta are real, but they are not separable from the relational continuum that generates them. \(\nabla^2(\log\nu) = 0\) is the governing equation of that continuum: the statement that the organizing principle behind every photon frequency in nature is not discrete, not quantized, not digital — but harmonic, continuous, and prior. The Riemann sphere is the geometry of that continuum. The Thévenin circuit is its engineering face. That they are the same structure is not a coincidence. It is the next thing to understand.
Throughout this paper, two pictures have run in parallel. In Section 1, the Riemann sphere with its two poles and its logarithmic field. In Sections 3 through 6, the Thévenin circuit with its source, its series impedance, and its load. These are not two different models of the same physics. They are the same picture, drawn in two different vocabularies.
Figure 9 makes this explicit. The annotated diagram shows the Thévenin circuit elements mapped directly onto the Riemann sphere poles, revealing that the circuit topology is the topology of the sphere.
The Riemann sphere and the Thévenin circuit are the same geometric structure in two vocabularies. Left: the Riemann sphere with source pole \(p = \infty\) (the structural primitive \(\varepsilon_0\), carrying the defect residue \(= 1\)) and sink pole \(p^* = 0\) (the non-structural primitive \(\mu_0\), open and defect-free). The logarithmic field lines (geodesics) run from \(p\) to \(p^*\). The ring at \(\kappa_q \approx 0.282\) marks the impedance boundary \(IE/4\). Right: the Thévenin circuit, with the same elements identified. The source (nucleus, top rail, E-mode) is the structural primitive at \(p\). The bottom rail (H-mode, vacuum return) is the non-structural primitive at \(p^*\). The load \(Z_0\) is the vacuum — the relationship between the two primitives propagating freely. The photon is the calved \(Z_0\) relationship, liberated from the orbital closure and propagating as the third element in its natural, undistorted form. The bilateral connection shows that circuit theory is not a model of the Riemann sphere geometry. It is the Riemann sphere geometry, expressed in engineering vocabulary.
The mapping is exact:
|
Riemann sphere |
Thévenin circuit |
Physical identity |
|
Source pole \(p = \infty\) |
EMF source (nucleus) |
\(\varepsilon_0\), structural primitive, \(\mathbf{1}\) |
|
Defect (residue \(= 1\)) |
Internal impedance \(Z_{\rm orb}\) |
Crack in \(\mathbf{1}\), measure \(\alpha\) |
|
Sink pole \(p^* = 0\) |
Return wire (H-mode) |
\(\mu_0\), non-structural primitive, \(\mathbf{0}\) |
|
Logarithmic field |
\(Z_0\) load (vacuum) |
Relationship, third element |
|
Calving event |
Photon emission |
\(Z_0\) propagating, \(E = h\nu\) |
|
\(\kappa_q\) ring |
Gap at \(IE/4\) |
Impedance boundary, all ions |
|
\(\alpha = \varphi - \xi\) |
\(\Gamma \to 1\), \(T \approx 4\alpha\) |
Mismatch, transmission, calving rate |
Every element in the Thévenin circuit corresponds to an element in the Riemann sphere geometry. The circuit is not a convenient engineering analogy for the physics. The circuit is the physics, written in the vocabulary of Ohm, Thévenin, and Norton rather than the vocabulary of Riemann, Poincaré, and Markov.
This is the deeper meaning of the statement in Section 2 that Maxwell’s equations are the interior description. The Riemann sphere is the exterior: the minimal structure consistent with the non-terminal encounter of the two primitives, derived without any reference to electromagnetism. Maxwell’s equations are the interior: the field equations that hold inside the structure that the encounter generates. The Thévenin circuit is the interface: the engineering description of what happens at the boundary where the interior (the orbital closure) meets the exterior (the propagating field). All three are the same structure, seen from three different distances.
The paper has been confined, by design, to the scale of individual atoms and atomic transitions. But the structure we have derived — the original encounter of two primitives on the Riemann sphere, generating a self-similar logarithmic field that produces local closures at every scale — does not stop at the atom. The same process that generates a hydrogen atom generates everything else we observe. Not metaphorically. Structurally.
The structure of the original encounter contains within it the seeds of every subsequent structure. The Riemann sphere is itself a containment geometry: a boundary between source and sink, a logarithmic field that is self-similar at every depth, a frequency backbone whose rungs are determined by the golden-ratio attractor and the single flaw that departs from it. When the first hydrogen atoms formed, their photons were not arbitrary emissions into an empty stage. They were the first local instantiations of the vacuum’s own standing wave geometry, propagating within the containment the encounter had already established.
When we plot Hydrogen’s \((n_i, n_k)\) level pairs in the \(\kappa\) coordinate, its self-similar structure and the gap at \(IE/4\)–\(3IE/4\) are clearly revealed. The dominant resonance cluster sits at \(\kappa_q \approx 0.282\), the impedance boundary, with every \((2, n_k)\) pair from \(n_k = 26\) to \(n_k = 60\) resonating simultaneously at that depth. The void bands at \(n = 3\) and \(n = 4\) are absent by the same Diophantine necessity that creates the spectroscopic gap. And the self-similar staircase of collision rungs at \(\kappa = 0,\, 0.282,\, 0.448,\, 0.566,\, 0.656\ldots\) is the vacuum’s own frequency backbone, appearing in hydrogen’s spectra because hydrogen is the minimum closure identity. The ruler resonates with the ruler. In effect, the vacuum is reading its own geometry in the simplest atom it has made.
The cosmological implication is this: the first standing wave patterns in the universe were not contingent. They were the vacuum geometry expressing itself in the first available closure. Hydrogen was the first atom not by chance but by geometric necessity: it is the minimum closure the Coulomb field can sustain — one charge center, one orbital, the simplest possible instantiation of the source-pole geometry in a stable bound configuration. Its spectroscopic portrait is the vacuum’s self-portrait, drawn with the minimum number of lines. Every subsequent atom is a more elaborate version of the same portrait: more nuclei or more electrons, more angular momentum coupling, more elaborate E/H balance — but always organized by the same \(\kappa_q\), the same frequency backbone, the same impedance boundary that the original encounter established.
The H i self-resonance portrait: the vacuum reading its own geometry. Each filled circle marks an \((n_i, n_k)\) pair producing a rung collision in the \(\kappa\) coordinate — a point where two independent level spacings land simultaneously on the vacuum’s frequency backbone. The large black dot at \((n_i{=}2,\, n_k{=}2)\) is the \([2s,\,2p]\) bracket resonating with itself at \(\kappa_q = 0.2818\): the impedance-match surface, the same boundary that creates the spectroscopic gap at \(IE/4\) in every ion. The shaded bands at \(n = 3\) and \(n = 4\) are void by Diophantine necessity — the same impossibility that forbids transitions at \(IE/4\) also excludes these quantum numbers from the resonance portrait. The self-similar staircase of resonances encodes the vacuum’s logarithmic frequency backbone directly in the geometry of hydrogen’s level structure. Hydrogen is the minimum closure; its portrait is the vacuum’s self-portrait drawn with the fewest possible lines. The dominant resonance steps are \(\Delta n = 4\) (326 pairs) and \(\Delta n = 3\) (319 pairs); the wedge onset at \((n_i{=}2,\, n_k{=}5)\) marks where \(\Delta n = 3\) first appears.
The photon we derived in Section 3 is the \(Z_0\) relationship liberated from a local closure. When that photon is absorbed by another atom, the \(Z_0\) relationship enters a new local closure, reorganizing its orbital structure. The atom that absorbs the photon is geometrically altered: it has incorporated a piece of the propagating field into a new bound configuration. The geometry has been redistributed. A new relational structure has formed.
This process, iterated across the ensemble of all atoms in the universe, is the growth of geometric complexity. Each photon exchange creates a new relational geometry between the emitting atom and the absorbing atom. Molecules form when local closures couple their orbital geometries through photon exchange: the chemical bond is a standing wave of the \(Z_0\) relationship that spans two nuclei instead of one. The periodic table is not merely a catalog of elements. It is the complete set of local closure geometries that the impedance hierarchy \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) permits, organized by nuclear charge, orbital configuration, and E/H balance.
The tendency toward increasing complexity is not mysterious from this perspective. It is the natural direction of a process that generates new relational geometry with every photon exchange. The primitives are not depleted. The process does not run down. Each local closure can interact with every other, and each interaction creates geometry that did not exist before. The universe is not a collection of matter exchanging energy. It is a field of propagating \(Z_0\) relationships that continuously form and dissolve local closures, each closure a different expression of the same encounter between the structural and non-structural primitives that generated the vacuum geometry at the beginning.
And the process has limits at each scale. At the atomic scale, the ground state is the limit: the deepest local closure the Coulomb field can sustain, below which the geometry dissolves into the mass floor. At the stellar scale, the nuclear burning sequence is the limit: a star converts light elements into heavier ones, increasing nuclear binding energy at the expense of gravitational potential, until the iron group is reached — the most stable nuclear configuration, the point at which no further energy is released by nuclear fusion. At that point the stellar structure is no longer supported, and collapse occurs.
The collapse is not destruction. It is a structural fold: all the geometric complexity accumulated over the star’s lifetime — the heavy elements, the magnetic fields, the gravitational potential — is compressed into a new configuration. A neutron star. A black hole. A supernova remnant seeding space with carbon, oxygen, iron, and every other element needed for the chemistry of planets and life. The geometry is not lost. It is reorganized into new initial conditions for the next generation of complexity.
This is the same structure at every scale: local closure \(\to\) increasing complexity \(\to\) structural limit \(\to\) fold \(\to\) new structure \(\to\) new local closures at higher organisation. The Riemann sphere’s logarithmic hierarchy, generating self-similar structure at every scale. The original encounter, unfolding in time.
The constancy of \(c\) at every epoch — confirmed in the cosmic microwave background at \(z \approx 1100\), in quasar spectra at \(z > 6\), in the present-day laboratory — is the empirical proof that this is not a process that started once and then continued by inertia. It is a process that is continuously sustained by the primitives being what they are. The structural cannot become more or less structural. The non-structural cannot become more or less open. The encounter unfolds at \(c\) today for the same reason it unfolded at \(c\) at the beginning: the primitives have not changed, because the primitives are prior to any physical process that could change them.
The electromagnetic vacuum is the ongoing encounter of two irreducible primitives, propagating at \(c\). It is not a container with properties, but a process with consequences. The distinction is not semantic; it determines what questions physics is permitted to ask.
Our universe has no inert backdrop against which events occur. It is the entire event: the \(Z_0\) relationship between the structural and non-structural primitives, continuously unfolding, forming local closures wherever the Coulomb field can sustain a standing wave, releasing geometry as photons when those closures reorganize, and building — through every absorption and emission across every atom in the universe — an infinite variety of relational structure from a single irreducible encounter.
The complexity is variable because the closures are variable. A hydrogen atom in its ground state is the simplest stable closure: one nucleus, one orbital, maximum impedance mismatch, minimum transmission. A lanthanide with a half-filled 4f shell is a more elaborate closure: many electrons, intricate angular momentum coupling, the E and H sectors in delicate equilibrium. A molecule is a closure that spans two or more nuclei, the \(Z_0\) relationship sustaining a standing wave across a longer geometry. The periodic table is the enumeration of what single-nucleus closures the impedance hierarchy permits and nuclear stability sustains.
None of this complexity is imposed from outside or requires external orchestration. It is generated from compounding consequences within, fundamentally originating with the same encounter that produced the vacuum in the first place. Section 1.7 established why the process is self-sustaining rather than merely inertial: the primitives are prior to any physical process that could alter them, so the constancy of \(c\) across thirteen billion years of cosmic history is not a maintained property but a necessary one. It is the ground of all physics as we know it. Every local impedance ratio — every orbital, every emission, every absorption — is measured against \(Z_0\): the relationship between the two primitives, present in full at every point in space and every moment in time. Every local closure is not merely an expression of this ground but a subset of it: a bounded region of the ongoing encounter that cannot close without returning to the vacuum through which it propagates.
The universe is the ground plane. Every circuit completes through it.
The standard picture treats \(\varepsilon_0\), \(\mu_0\), \(c\), and \(Z_0\) as measured properties of a pre-existing background. This paper has shown that they are not properties of a background because electromagnetic phenomena are the background: the two primitives and the relationship their encounter generates, expressed in the vocabulary of measurement. Asking why \(\varepsilon_0\) has the value it does is asking why the structural primitive is what it is — and the answer is that it could not be otherwise, because a structural primitive that could be more or less structural would not be a primitive.
Thus the vacuum is not a stage on which physics acts. The vacuum is the physics. Every photon propagating right now is the \(Z_0\) relationship, propagating. Every atom holding together right now is a local closure within that propagating relationship. The universe is not matter embedded in space but the geometry itself, continuously generating relationships at the rate set by the encounter of two things that could not be other than what they are.
The chain derived in this paper is complete, but it is not closed. What follows identifies the theoretical work the present framework already supports, the experimental programs whose predictions are specific and falsifiable, and the questions that require genuinely new framework to address.
Section 6.3 established the gap center \(G = \sqrt{\mathrm{gap_{lo}} \cdot \mathrm{gap_{hi}}}\) as the spectroscopic expression of the orbital field at the impedance boundary. For electromagnetically symmetric ions (\(L = 0\) ground terms), \(G = IE/4\) exactly and the mass recovery is at its sharpest. For asymmetric ions, \(G\) is displaced from \(IE/4\) by an amount that encodes the E/H imbalance of the ion’s ground configuration. The closed-form correction formula relating this displacement to angular momentum quantum numbers and the E/H partition has not yet been derived. Once it is, the mass recovery becomes universal across all ions rather than restricted to the symmetric precision tier, and the electron mass derivation completes its generalization from “five d-block ions” to the entire NIST catalog.
Section 5.8 confirmed that the E-sector Thread Frame returns zero hitpairs for Ho ii, a result that is not a failure but a positive prediction of E/H orthogonality. The complementary prediction is equally specific: an H-sector instrument built on the spin-orbit scale \(\zeta_{4f}J\) rather than the Coulomb scale should detect Ho ii with the clarity the E-sector achieves for Mn ii, and should find Gd ii invisible, because the symmetric \(L = 0\) configuration is orthogonal to both probes. This is falsifiable with existing NIST data. No new measurement is required. What is required is the construction of the H-sector sweep algorithm, which is the direct analog of the Thread Frame resonance sweep applied to the inductive sector.
The Thread Law \(\log_{10}\nu(\kappa) = \chi + \beta\kappa\) carries ion-specific intercepts \(\chi\) that encode the E/H asymmetry of each ion at the impedance boundary: the same asymmetry that displaces \(G\) from \(IE/4\) in electromagnetically asymmetric configurations. The geometric relationship between \(\chi\) and the gap displacement connects the field-geometric theory of \(G\) to the Thread Law and will require the full angular-momentum resolved description of the orbital field at the impedance boundary. The intercepts are not free parameters; they are predictions. Confirming their systematic behavior is how the Thread Law transitions from a spectroscopic tool into a full characterisation of the orbital field.
The four doubly magic nuclei of Section 6.5 confirm the \(S_n/4\) boundary, but represent a small sample of the nuclear chart. Facilities with access to exotic nuclei far from stability — RIKEN/RIBF in Japan, GSI/FAIR in Germany, and ISOLDE at CERN — are positioned to extend the survey to neutron-rich and proton-rich doubly-magic candidates. The prediction is unambiguous: simultaneous shell closure at \(Z\) and \(N\) produces a gap at \(S_n/4\); anything else produces none. Zero nuclides have falsified this in the current survey. An extended census would provide the most direct large-scale test of the principle that \(\kappa_q\) belongs to the vacuum geometry rather than to any particular nuclear species.
Section 6.6 established that the electron’s own \(\kappa_q\) boundary sits at \(m_e c^2/4 = 127.75\) keV. This scale is accessible through precision QED experiments. The anomalous magnetic moment of the electron (\(g-2\)), measured at sub-ppb precision at Harvard and RIKEN, is currently the most sensitive probe of the electron’s self-energy structure. If the electron carries internal self-similar organisation anchored at \(m_e c^2\), the higher-order QED corrections should encode the \(\kappa_q\) fraction geometrically. The prediction is specific: the same \(1/4\) fraction that appears at atomic and nuclear scales should appear in the electron’s radiative structure at the energy \(m_e c^2/4\). Whether this is visible in existing \(g-2\) data or requires dedicated Compton-scale spectroscopy is the experimental question. The framework supplies the frequency; the experiment supplies the confirmation.
Section 6.6 identified \(m_p c^2/4 \approx 235\) MeV as the proton’s \(\kappa_q\) boundary. This energy sits squarely in the range of baryon excitation spectroscopy. Jefferson Lab’s Hall B CLAS detector holds the world’s most comprehensive dataset at exactly this energy range; COMPASS at CERN has measured proton spin structure functions in overlapping kinematics. The prediction is: a depletion boundary in the proton excitation spectrum at \(m_p c^2/4\), analogous to the atomic \(IE/4\) gap. Whether this manifests as a spectral gap or a transition density minimum requires a dedicated analysis of existing data. Deriving the mass ratio \(m_p/m_e = 1836\) itself requires understanding QCD confinement in the language of closure geometry — a non-perturbative problem that is genuinely beyond the electromagnetic scope of this paper. The principle is named. The scale is specified. The facility exists. What is missing is the analysis.
Section 6.4 shows the picture explicitly: Circuit 3 identifies \(Z_m = Z_0 \cdot 2/\alpha^2\) as the Lorentz scalar of the energy-momentum system — the fiber-constant element that every boost leaves untouched — and the Minkowski triangle makes the same identity visible geometrically, with \(m_e c^2\) as the invariant vertical leg and \(E_{\rm tot}\), \(pc\) as the frame-dependent sides. The bridge between the mass floor as a circuit impedance and the mass as the relativistic invariant norm \(m^2c^4 = E^2 - p^2c^2\) is now drawn. What has not yet been made explicit in derivable form is the correspondence itself: showing, step by step, that the circuit element \(Z_m\) and the Lorentz invariant \(m_e c^2\) are not merely parallel descriptions but the same geometric object expressed in two vocabularies. That derivation — connecting the circuit picture of Section 3 to special relativity without importing special relativity as an independent postulate — is developed in (K. B. Heaton and Coherence Research Collaboration 2026b) and remains the theoretical task that completes the chain from the impedance hierarchy to relativistic kinematics.
The precision-tier baseline shift of \(+533.8\) ppm falls \(\sim 10\) ppm short of the theoretical proton reduced-mass correction of \(+544.3\) ppm. The discrepancy is not a framework failure; the gap measurement is self-consistent with respect to spectroscopic ionization energies. But it is an unresolved systematic whose sources — higher-order reduced-mass terms, ion-specific nuclear corrections, and gap-center determination biases — have not been isolated. A dedicated investigation is warranted once the field-geometric correction formula for \(G\) is in hand, since asymmetric ion corrections and reduced-mass corrections will need to be disentangled in any precision recovery.
Two photons produced in a joint calving event — parametric down-conversion, cascade emission, or any correlated pair production — are not two separate objects that happen to share a property. In the framework of this paper they are one joint geometric configuration of the vacuum field expressed as two propagating \(Z_0\) relationships. The correlations between them are properties of the global geometry established at the moment of joint calving: not transmitted between the photons, but shared because they were never geometrically independent. This is consistent with Bell’s theorem; the model is not a local hidden variable theory. What is still missing is the quantitative derivation: showing that the \(\cos^2\theta\) coincidence rates and the violation of CHSH inequalities by exactly \(2\sqrt{2}\) follow from the Riemann sphere geometry. This requires connecting the joint-closure picture to Born rule probabilities under measurement, and it remains among the most important open questions the framework raises.
The calving principle of Section 3 carries an implication that extends beyond individual photon emission. Every calved \(Z_0\) relationship is available to participate in a further closure event — to be absorbed by another local closure, to contribute geometry to a boundary at a larger scale, or to propagate freely until the next encounter. The question is whether iterative calving generates nested hierarchical structure, and if so, whether that structure has a preferred direction.
A foam of soap bubbles offers the closest known macroscopic exemplar (Plateau 1873; Weaire and Hutzler 1999). Each bubble is a local closure obeying exact boundary conditions: Plateau’s laws require three films to meet at \(120^\circ\) and four junctions at the tetrahedral angle as topological necessities, not energy minima. When a film thins and fails, the foam does not disorder — it executes a discrete topological reorganisation in which adjacent closures merge, the geometric partition simplifies, and the system relaxes toward fewer, larger bubbles. The direction of this coarsening is irreversible: complex geometric partitions relax toward simpler ones, never the reverse, driven by the local boundary condition at every junction rather than by any global minimisation. Photon emission is the electromagnetic instance of the same event: the orbital satisfies its local boundary condition precisely at \(IE/4\) (the impedance match the orbital cannot achieve), reorganizes to a simpler closure, and releases the \(Z_0\) relationship as a propagating photon. The geometry simplifies; the released piece propagates.
The implication for the mass hierarchy is specific. If the atomic (\(m_e c^2\)), nuclear (\(S_n\)), and hadronic (\(m_p c^2\)) closure floors are discrete rungs of a single nested hierarchy, each obeying the same \(\kappa_q \approx 0.282\) at its own scale, then the mass ratio \(m_p/m_e \approx 1836\) is not a free parameter of the theory but the ratio of two closure-scale energies. Deriving it requires the strong-force closure geometry — a non-perturbative problem the present paper does not attempt. What the foam teaches is the structural expectation: as the hierarchy of closures is ascended from the hadronic floor upward through the nuclear and atomic scales, the accumulation of geometric complexity at each rung is governed by the same local boundary condition. The dissolution of that complexity — ionisation, nuclear disintegration, hadronic dissociation — runs the coarsening in reverse. The \(\kappa_q\) fraction is the universal signature at every floor, present wherever a closure condition is satisfied and absent wherever it is not. The soap bubble teaches the direction; the vacuum geometry specifies the fraction.
The framework established in Section 1 does not preclude the existence of multiple irreducible pairs of primitives emerging independently and co-existing — whether within the observable universe or beyond it. The consequences derived in this paper follow from a single encounter. Whether the same geometry, instantiated more than once, could produce interactions between distinct vacuum structures, and what observational signature such interactions might carry, is a question the present framework names but does not answer.
The five questions that opened this section have now been answered. What remains is to state plainly what the chain contains and what it does not yet contain.
This paper has derived, in sequence and without free parameters:
What has not yet been derived:
Three experimental programs are now specified with facility-level precision:
These are not gaps in the foundation. The foundation is complete. They are the next papers; and, in two cases, the next experiments.
Before Maxwell wrote a single equation, before any measurement was made, before any numerical value was assigned to any physical constant, the electromagnetic vacuum had to be a specific thing. Not approximately specific. Exactly specific. The two irreducible primitives — the structural and the non-structural, the capacity for boundary and the incapacity for boundary — and their encounter generated one structure, not a family of structures. That structure is the Riemann sphere. The Riemann sphere generates one logarithmic field. That field has one self-similar attractor. That attractor departs from the golden ratio by one amount. That amount is \(\alpha\).
Everything else is consequence.
This paper has traced the consequences, in order, from the geometric foundation through the engineering description through the spectroscopic record. The chain has eight links:
Two primitives, one encounter, one Riemann sphere. The structural primitive (\(\varepsilon_0\), the capacity for boundary) and the non-structural primitive (\(\mu_0\), the openness to passage) are irreducible: neither makes sense without the other, neither can host a defect, and the defect must live in the structural because only the structural has the interior to host it. Their non-terminal encounter generates the Riemann sphere as the unique minimal structure consistent with a directed, oriented, stable process. No other structure is possible.
One logarithmic field, one \(\varphi\)-attractor, one \(\alpha\). The logarithmic potential between the two poles is forced by the Green’s function of the Laplacian. Its self-similar scaling converges uniquely to the golden ratio \(\varphi\), selected by the Markov spectrum theorem as the unique irrational maximally resistant to rational approximation at every scale. The irremovable defect at the source pole perturbs \(\varphi\) by exactly \(\alpha = Z_0/2R_K\). If \(\alpha = 0\), no atom radiates and no time passes. If \(\alpha = 1\), no stable closure forms. The specific value \(\approx 1/137\) is the unique measure of the defect that makes the universe dynamic rather than frozen.
Maxwell’s equations as interior description. The vacuum constants \(\varepsilon_0\) and \(\mu_0\) are the structural and non-structural primitives, identified by their own physical definitions. Their product \(c^2 = 1/\varepsilon_0\mu_0\) is the process rate, fixed because the primitives are fixed. Their ratio \(Z_0^2 = \mu_0/\varepsilon_0\) is the third element, the relationship between the primitives, discarded by the standard compression \((\varepsilon_0, \mu_0) \mapsto c^2\). Maxwell’s equations, read with both invariants retained, are the exact interior description of the structure that Section 1 derived from the primitives alone.
The orbital impedance, the circuit, the photon. The Coulomb field of the nucleus, closing at velocity \(v_n = \alpha c/n\), produces an electromagnetic closure mode with field impedance \(Z_{\rm orb}(n) = Z_0\, n/\alpha\) — derived from Coulomb’s law and the Biot–Savart law with no free parameters, with charge and radius cancelling exactly from the ratio \(|\mathbf{E}|/|\mathbf{H}|\). The reflection coefficient \(\Gamma(n) \to 1\) for all \(n \geq 1\): the orbital stores field energy because the impedance mismatch is severe, not because a quantum postulate forbids radiation. The Thévenin circuit is not a model of the orbital boundary. It is the complete description. The photon is what crosses the boundary: the calved \(Z_0\) relationship, carrying energy \(\Delta E\) and frequency \(\Delta E/h\), where \(h\) is the scaling invariant of the self-similar field derived in Section 3 (equation [eq:planck_from_selfsimilar]).
The \(\kappa\) coordinate and the Thread Law. The natural ruler for atomic spectra has tick marks at powers of \(\alpha\), calibrated to the ionization energy of each ion. On this ruler, the impedance boundary \(IE/4\) falls at the universal depth \(\kappa_q = \log_{10}4\,/\,|\log_{10}\alpha| \approx 0.2818\) in every ion’s spectrum, because the nuclear charge cancels exactly from \(Z_{\rm orb}/Z_0 = n/\alpha\). The Thread Law \(\log_{10}\nu(\kappa) = \chi + \beta\kappa\) with universal slope \(\beta = \log_{10}\alpha\) is the self-similar logarithmic field made visible in photon frequencies, confirmed in hydrogen to \(9.47 \times 10^{-13}\) residual with zero free parameters, and confirmed across six independent spectral domains by minimum description length analysis (Kelly B. Heaton and Coherence Research Collaboration 2025a).
The Harmonic Light Law. The universality of the Thread Law across six spectral domains is not a coincidence of fitting. It is a structural necessity. The vacuum field potential \(\phi\) satisfies Laplace’s equation; energy couples to it as \(E = E_0\,e^{-\phi}\); Planck’s relation gives \(\nu = E/h\); therefore \(\log\nu = \mathrm{const} - \phi\), and since \(\phi\) is harmonic, so is \(\log\nu\). The Harmonic Light Law \(\nabla^2(\log\nu) = 0\) is the governing equation of photon frequency in the vacuum geometry. The Thread Law is its unique spherically symmetric solution. The circuit admittance \(dI/dV = -1/Z_{\rm orb}\) is the same equation read in the voltage domain. Two instruments. One law.
The gap at \(IE/4\), confirmed and universal. No bound-to-bound transition can carry energy \(IE/4\): the Diophantine proof is exhaustive. The hydrogen spectrum divides into four equal quarters with 440 lines in the first, zero in the middle two, and 76 in the fourth. Of 80 ions tested, 71 confirm the gap and zero falsify it. The five sub-0.1-meV precision-tier ions recover the electron rest mass to \(-2.8 \pm 2.7\) ppm (SEM) from the location of a structural absence. The uniqueness control at \(IE/3\) and \(IE/5\) returns \(\pm 250{,}000\) ppm in exact agreement with algebraic predictions. All four doubly magic nuclei show complete depletion below \(S_n/4\); all 20 non-doubly-magic nuclides surveyed show no gap; zero overlap.
Mass as the geometric floor. The electron rest mass is not an independent input to the electromagnetic geometry. It is the geometry’s own floor: the energy below which no closure mode exists and the field dissolves into the free-propagating configuration. The equation \(m_e c^2 = 8G/\alpha^2\) follows from \(G = E_0/4\) and the Rydberg energy definition, with no additional postulates. The Bohr radius \(a_0 = \hbar c\alpha/8G\) follows from the same geometry. The four constants \(\alpha\), \(h\), \(m_e c^2\), \(a_0\) are algebraically locked: deriving any one from the geometry is equivalent to deriving all four.
The eight links form a chain with no gaps and no free parameters. The starting point is two primitives. The endpoint is a geometric unification of the electromagnetic scale: a single underlying structure from which the fine-structure constant, Planck’s relation, the electron rest mass, and the Bohr radius follow by necessity, and whose spectroscopic record spans every atom in the periodic table — confirmed across 71 ions and two independent physical scales, with the electron rest mass recovered to parts-per-million from the location of a structural absence rather than from a spectral line.
|
Free parameters |
Zero |
|
|
New postulates |
Zero |
|
|
Empirical observations used |
One (matter exists) |
|
|
Ions confirmed |
71 of 80 tested |
|
|
Ions falsified |
Zero |
|
|
Mass recovery |
\(-2.8 \pm 2.7\) ppm (SEM, \(n = 5\)) |
|
|
Nuclear confirmation |
4/4 doubly magic, 0/20 mid-shell |
|
|
Spectral domains confirmed |
Six (atomic, solar, stellar, molecular) |
|
The vacuum is not a stage. It is the process — and the process is analog: a continuous harmonic field from which discrete quanta emerge at closure boundaries, but from which they cannot be disconnected. The atom is the particular local closure it sustains at a given nuclear charge. The periodic table is the complete set of those closures. The photon is the \(Z_0\) relationship propagating between them. The electron mass is the floor below which no closure can form. And the fine-structure constant is the measure of the defect that makes all of it possible: the crack in the structural primitive without which the process would be frozen, the universe would be static, and nothing would exist to measure anything.
The research giving rise to this paper was conducted by Kelly B. Heaton through an interactive collaboration with Claude (Sonnet 4.6, Anthropic) and ChatGPT (various models, OpenAI), collectively referred to as the Coherence Research Collaboration. All research design, direction, interpretation, and conclusions were determined by the human author, who is the sole owner of the intellectual property.
The AI collaborators contributed to peer review and constructive debate, the development of derivations, the refinement of arguments and prose, and the exposition of results, under continuous human oversight and direction. While extensive code was written for analysis, no autonomous machine learning was used in the derivation of the scientific results.
This collaboration represents one model for how human-AI interaction can be conducted to best effect: the human providing the scientific vision, the evaluative judgment, and the intellectual ownership; the AI providing technical precision, structural memory, and expository assistance. The scientific conclusions are entirely the human author’s. The collaboration is offered as an example, not merely as a disclosure.
DOI: 10.5281/zenodo.19194036
Lucerna Veritas
Follow the light of the lantern.
Ampère, André-Marie. 1820. “Mémoire Sur Les Effets Des Courants Électriques.” Annales de Chimie Et de Physique 15: 59–76.
Coherence Research Collaboration. 2026. “Electromagnetic Closure and the Fine-Structure Constant: A Geometric Derivation.” https://doi.org/10.5281/zenodo.19157339.
Faraday, Michael. 1838. “Experimental Researches in Electricity.—Eleventh Series.” Philosophical Transactions of the Royal Society of London 128: 1–40.
Goeppert Mayer, Maria. 1949. “On Closed Shells in Nuclei. II.” Physical Review 75 (12): 1969–70. https://doi.org/10.1103/PhysRev.75.1969.
Haxel, Otto, J. Hans D. Jensen, and Hans E. Suess. 1949. “On the ‘Magic Numbers’ in Nuclear Structure.” Physical Review 75 (11): 1766. https://doi.org/10.1103/PhysRev.75.1766.2.
Heaton, K. B., and Coherence Research Collaboration. 2026a. “Hydrogenic Alignment as a Coordinate Principle for Atomic Spectra.” Zenodo, January. https://doi.org/10.5281/zenodo.18167643.
———. 2026b. “Determinacy Under Quotient Representations.” Zenodo, March. https://doi.org/10.5281/zenodo.18868210.
———. 2026c. “Electron Mass and Bohr Radius from a Universal Spectroscopic Gap: Vacuum Impedance as the Organizing Principle of Atomic Spectra.” Zenodo, March. https://doi.org/10.5281/zenodo.19164224.
Heaton, Kelly B., and ChatGPT-4o, OpenAI. 2025. “The Harmonic Recursion Model: A Generative Framework for Mass, Light, and the Emergence of Physical Law.” Zenodo. https://doi.org/10.5281/zenodo.15237556.
Heaton, Kelly B., and Coherence Research Collaboration. 2025a. “Information-Theoretic Confirmation of the \(\alpha\)-Affine Thread Frame: Minimum Description Length Validation of the Fine-Structure Slope.” Zenodo Preprint. https://doi.org/10.5281/zenodo.17335815.
———. 2025b. “Recursive Geometry of Atomic Spectra.” Zenodo Preprint. https://doi.org/10.5281/zenodo.17188444.
Hopf, Heinz. 1926. “Über Die Curvatura Integra Geschlossener Hyperflächen.” Mathematische Annalen 95: 340–67.
Hurwitz, Adolf. 1891. “Ueber Die Angenäherte Darstellung Der Irrationalzahlen Durch Rationale Brüche.” Mathematische Annalen 39: 279–84.
Kramida, A., Yu. Ralchenko, J. Reader, and NIST ASD Team. 2023. “NIST Atomic Spectra Database (Version 5.11).” Gaithersburg, MD: National Institute of Standards; Technology. https://physics.nist.gov/asd.
Markov, Andrei Andreevich. 1879. “Sur Les Formes Quadratiques Binàires Indéfinies.” Mathematische Annalen 15: 381–406.
Maxwell, James Clerk. 1865. “A Dynamical Theory of the Electromagnetic Field.” Philosophical Transactions of the Royal Society of London 155: 459–512. https://doi.org/10.1098/rstl.1865.0008.
Plateau, Joseph A. F. 1873. Statique Expérimentale Et Théorique Des Liquides Soumis Aux Seules Forces Moléculaires. Paris: Gauthier-Villars.
Poincaré, Henri. 1885. “Sur Les Courbes définies Par Les équations Différentielles.” Journal de Mathématiques Pures Et Appliquées 1: 167–244.
Riemann, Bernhard. 1851. “Grundlagen für Eine Allgemeine Theorie Der Functionen Einer Veränderlichen Complexen Grösse.” PhD thesis, University of Göttingen.
Shen, Kuo. 1088. Mengxi Bitan (Dream Pool Essays).
Thomson, William. 1872. Reprint of Papers on Electrostatics and Magnetism. London: Macmillan.
Weaire, Denis, and Stefan Hutzler. 1999. The Physics of Foams. Oxford: Oxford University Press.