# 8. Standing Waves and Discreteness Chapter 7 established that source-free energy flow satisfies the wave equation $$ \partial_t^2 \mathbf F-c^2\nabla^2\mathbf F=0, \qquad \nabla\cdot\mathbf F=0. $$ As a useful representation, each Cartesian component therefore satisfies $$ \left(\nabla^2-\frac{1}{c^2}\partial_t^2\right)f=0. $$ Chapter 7b derived the self-refraction principle: retarded portions of the same flow alter the local transport law and bend later transport of that same field. This chapter asks what follows once that bending becomes strong enough to make the path close on itself. Toroidal configurations are not exotic. They arise spontaneously whenever a flow curls back on itself — smoke rings, vortex rings in water, incense plumes disturbed by a hand. What is common to all these cases is that toroidal geometry forms easily; what distinguishes them is whether the resulting closure can self-sustain. Most such configurations dissipate. The ones that persist are those whose self-refraction energy density matches the geometric curvature needed to maintain the closure (Appendix A0). This chapter does not address the self-consistency of the closure. It asks only what standing-wave organizations are allowed once a toroidal closure exists. A self-refracting closure must be a shape that admits continuous nowhere-vanishing tangential flow. A sphere does not: by the hairy ball theorem, no continuous nowhere-vanishing tangential vector field exists on a sphere. The simplest closed shape that can sustain such flow is a torus, a sphere with a smooth through-hole. It has two independent non-contractible cycles, and the flow must close in both directions at the same time. As we shall see, a self-refracting flow closing toroidally yields integer modes and the Rydberg-type $1/n^2$ scaling. Since hydrogen is matter, this is a first serious clue that matter itself may be organized self-refracting closures of energy flow. For modes organized along the toroidal closure, those whose dominant structure wraps the two cycles, the natural starting point is not a fixed Cartesian component and not even fixed toroidal angles. Start with a closed centerline $\gamma(s)$ parametrized by arclength $s\in[0,L)$, and around it choose a local frame in which $\hat{\mathbf t}$ is tangent to the centerline and $\hat{\boldsymbol\theta}$ is the turning direction around the local cross-section. The transporting direction is then a helical tangent $$ \hat{\mathbf f} = \cos\beta\,\hat{\mathbf t} + \sin\beta\,\hat{\boldsymbol\theta}. $$ Here $\beta$ is the local winding angle, measured from the centerline tangent. Chapter 7b already derived that local self-refraction fixes the bending angle through $$ \cos\beta=\frac{1}{n_{\mathrm{eff}}}, \qquad \tan\beta=\sqrt{n_{\mathrm{eff}}^2-1}. $$ This chapter needs only the geometric consequence: once self-refraction has produced a toroidal closure, the winding can be treated as approximately uniform. The self-consistent aspect ratio of such a closure is derived in Appendix A0. The mode counting below does not depend on that ratio: the integer conditions are topological and hold for any aspect ratio. The separated wave equation used below omits curvature corrections that depend on the aspect ratio; these corrections shift the exact mode frequencies but do not change the integer counting (see Appendix A2 for the full toroidal Laplacian). Take a toroidal closure with cross-sectional radius $r$ and centerline length $L=2\pi R$, with the winding angle $\beta$ approximately constant along a closed streamline. If one such streamline winds $m$ times around the major cycle and $n$ times around the minor cycle before returning to itself, then $$ \Delta s = mL, \qquad \Delta\theta = 2\pi n. $$ Since $\tan\beta$ is the ratio of transverse advance to longitudinal advance, $$ \tan\beta = \frac{r\,\Delta\theta}{\Delta s} = \frac{nr}{mR}. $$ So closure of a single helical streamline is exactly the statement that its slope is rational: it returns only after integer counts on the two non-contractible cycles. Combining the geometric closure with the constitutive refraction relation gives $$ \tan\beta=\frac{nr}{mR}=\sqrt{n_{\mathrm{eff}}^2-1}, $$ so equivalently $$ \beta = \arctan\!\left(\frac{nr}{mR}\right) = \arccos\!\left(\frac{1}{n_{\mathrm{eff}}}\right), $$ and $$ n_{\mathrm{eff}} = \sqrt{1+\left(\frac{nr}{mR}\right)^2}. $$ The standing field is a stronger condition than closure of one streamline. It must be single-valued on both cycles of the toroidal closure. Resolve the field by a scalar amplitude $f(s,\theta,t)$ in this moving frame. Then, to leading order, $$ \partial_t^2 f = c^2\left( \partial_s^2 f + \frac{1}{r^2}\partial_\theta^2 f \right), $$ with curvature corrections omitted. These corrections, which depend on the aspect ratio $r/R$, shift the exact mode frequencies but do not affect the integer counting. The full toroidal wave equation with all curvature terms retained is treated in Appendix A2. Because the closure is self-consistent, the field must be periodic on both cycles: $$ f(s+L,\theta,t)=f(s,\theta,t), \qquad f(s,\theta+2\pi,t)=f(s,\theta,t). $$ Now seek a separated standing mode $$ f(s,\theta,t)=A\cos(ks)\cos(n\theta)\cos(\omega t). $$ The $\theta$-periodicity forces $$ n\in\mathbb Z_{\ge 0}, $$ while the $s$-periodicity forces $$ kL=2\pi m, \qquad m\in\mathbb Z_{\ge 0}. $$ So the same pair of integers appears twice: first as the winding counts of a closed helical streamline, and second as the phase counts of a standing field that is single-valued on the two fundamental cycles. Since $L=2\pi R$, this gives $$ k=\frac{m}{R}. $$ Substituting the standing mode into the wave equation yields $$ \omega^2 = c^2\left(k^2+\frac{n^2}{r^2}\right) = c^2\left(\frac{m^2}{R^2}+\frac{n^2}{r^2}\right). $$ So the torus discretizes the transport immediately. The closed geometry permits only integer mode numbers and therefore only discrete standing-wave frequencies. If one prefers the more familiar angular coordinate around the major cycle, $$ \phi = \frac{s}{R}, $$ then the same leading-order closure equation becomes $$ \partial_t^2 f = c^2\left( \frac{1}{R^2}\partial_\phi^2 f + \frac{1}{r^2}\partial_\theta^2 f \right). $$ The same result can be written as closure in wavelength form: $$ m\lambda_s = L = 2\pi R, \qquad n\lambda_\theta = 2\pi r, $$ for integers $m,n\in\mathbb Z_{>0}$. Equivalently, $$ k_s=\frac{m}{R}, \qquad k_\theta=\frac{n}{r}. $$ So a bounded toroidal standing wave is not labeled by a continuous parameter, but by an integer pair $(m,n)$, and its frequency is $$ \omega_{mn} = c\sqrt{k_s^2+k_\theta^2} = c\sqrt{\frac{m^2}{R^2}+\frac{n^2}{r^2}}. $$ So discreteness enters before any particle picture. Once the field is required to close on itself in a toroidal closure, only certain standing-wave organizations are allowed. This is the right way to read the early quantum fact that hydrogen radiates in discrete lines. The discreteness does not require an electron moving on planet-like orbits. It requires only that bounded energy flow reorganize itself between allowed standing-wave closures. The observed Rydberg pattern can then be read as a special family of such reorganizations. If a fixed toroidal closure is refined into an $N\times N$ standing-wave partition, the same total energy is distributed across $N^2$ coherent cells. The characteristic energy per cell therefore scales as $$ E_N \propto \frac{E_1}{N^2}. $$ Transitions between two such allowed organizations then have the form $$ \Delta E \propto \frac{1}{p^2}-\frac{1}{q^2}, \qquad p>q. $$ The integers are not mysterious labels imposed from outside. They are the counting numbers of the standing-wave closure itself. So discreteness begins as standing-wave closure of source-free energy flow in a self-refracting toroidal configuration. Once that closure exists, its further global aspects can be separated. The narrow-band envelope sector of the same bounded mode appears as Schrodinger dynamics in the next chapter. The through-hole character of that same toroidal standing wave appears as charge in the chapter after that.