VANIAN TESSERACTICS

Tesseractions

Tilings for Non-uniform Denominators

Dr. L. Van Warren - Original Idea
Kimi K3, Gemini Flash 3.6 - Assistance with Mathematical Development

Being a Volume of the AddSubMulDivia Series,
with Selections from the Companion Series IntDiffLogExpia

An Interactive HTML5 Edition · Set by the Press
Requires an internet connection for mathematical typesetting (MathJax).

❦

Dedication

To the mosaicists of Rome, who first made wholes from unequal tiles; to Pythagoras, who held the four sacred; to Hinton, who named the fourth dimension; to Eudoxus and Archimedes, who exhausted what they could not count; and to Van, who asked why the pieces had to be equal.

Canonical Glossary & Notation

The Press sets forth the settled terminology. The objects are unchanged throughout; only the tiles vary.

Term	Meaning
tesseraction	a part-against-whole object $\tes{A}{\Omega}{\rho}$; an act of tiling
tessera	one weighted piece of an opus
opus	the whole $\Omega$, together with its tiling rule
tiling rule	the weight assignment $w$ (discrete) or density $\rho$ (continuous)
vermiculation	refinement: passage to finer tesserae
equitesseral	equal-tiled; the classical fractions
proper / improper	value in $[0,1]$ / value outside it
pooling	the mediant sum of two tesseractions over juxtaposed opuses
ascent	iteration of a tesseraction against itself: powers
tesseractics	the discipline
Vanian	of the founder's school

The Settled Hand (Notation). A tesseraction is an ordinary fraction wearing only the one extra fact an uneven whole requires — the rule: $$T \;=\; \tes{A}{\Omega}{\rho}, \qquad\text{“the share of }A\text{ in the opus }\Omega\text{, tiled by }\rho\text{.”}$$ Its value is $$\tval{A}{\Omega}{\rho} \;:=\; \frac{\mu_\rho(A)}{\mu_\rho(\Omega)},$$ where in the discrete world $\Omega=(I,w)$, $\mu_w(A)=w(A)=\sum_{i\in A}w_i$, $W=w(I)$; in the continuous world $\mu_\rho(A)=\int_A \rho\,dx$. When the rule is constant and the tiles equal, this degenerates to the familiar $\frac{a}{b}$ — the equitesseral check every theorem in this book must pass (and does). Throughout, $|T|$ denotes value; bold $\mathbf{v}$ denotes velocity; no symbol serves two masters.

Section Index & Enhancement Map

Every chapter and widget carries a stable HTML id. Cite the id when requesting revisions.

id	Contents	Widgets
`ch01`	Founding question; where tesseractions already live	—
`ch02`	Foundations; Theorem of Equivalent Shares; product & coproduct	`w01` builder, `w02` vermiculation, `w03` mirror
`ch03`	Addition; inclusion–exclusion; pooling; Simpson	`w04`, `w05`
`ch04`	Subtraction; signed shares; Jordan decomposition	`w06`
`ch05`	Multiplication; Fubini; telescoping	`w07`, `w08`
`ch06`	Division; reciprocal swap; quotitive; mixed shares	`w09`
`ch07`	Powers; Tesseract Theorem; chambers; roots	`w10`, `w11`
`ch08`	Axiom of Approach; infinite ascent; logarithms	`w12`, `w13`
`ch09`	The natural rung; $e$; the gauge	`w14`
`ch10`	Differentiation of tesseractions	`w15`
`ch11`	Accumulation / occupancy	—
`ch12`	Exponential morphing; information	`w16`
`ch13`	Equi-distribution; CFD grids; ALE; airfoil meshing	`w17`, `w18`
`ch14`–`ch17`	Forms, quantum/stochastic, relativity, categories	hooks reserved `w19+`
`bm-*`	Closure tables, named theorems, errata, index, reading	—

Every widget wNN has a toggleable Python companion pyNN immediately beneath it — self-contained numpy/matplotlib source, with a Colab jump-out badge and a zero-setup fallback (paste into colab.new). The companions reproduce the widget's theorem in the Press palette; the motivated reader runs them as-is.

Part 0 · The Question and the Landscape

Chapter 1 · The Founding Question

So I was recently reviewing fractions, and the denominator of a fraction is the number of equal-sized pieces a (virtual?) original whole is divided into, compared with the numerator, which determines the value of the entire fractional expression. So then I thought: what if we had a system where the original whole is divided into some number of variable-sized pieces, and out of necessity one provides a rule, function, or relation on how that virtual original whole would be subdivided, that for utility would also give you the size of the particular piece you were dealing with? This would result in the emergence of a whole logical system of reasoning, along with its cognate notation, that one could derive statements of truth, or theorems about.

One could specify the puzzle-piece in question as simply the integral of some specific rule with limits as to where the boundary of the segment began and ended. For rules that partition the denominator's original whole in one dimension, one would need only a one-dimensional rule, along with the limits of integration, to get a definite result. We could also divide the denominator in two directions, in which case we would use the definite double integral, with two sets of limits, and so on, partitioning in as many dimensions as we would like.

Has any such number and notation system already been formalized? If so, what is its utility — and if not, how might we design one?

— the founding question & pun on definite results

§1.1 · The Press Responds: Where Tesseractions Already Live

Short answer: pieces of this have been formalized many times, and the grand unifying version is one of the crown jewels of twentieth-century mathematics — measure theory, with probability theory as its normalized special case. The founding vocabulary translates cleanly:

Vanian term	Standard term
virtual original whole	a measure space $X$
rule for subdividing	a measure $\mu$, or a density $\rho$ with $\mu(A)=\int_A\rho$
method of specifying the piece	a measurable subset $A$ (or an index set, discretely)
the tesseraction's value	the share $\mu(A)/\mu(X) = \int_A\rho \;/\; \int_X\rho$

One key reframing the question forces: in an ordinary fraction the numerator is just a count — “how many pieces” — which only works because all pieces are identical. Once pieces vary, the numerator must generalize from a count to a selection: which pieces. That single move is basically the birth of measure theory.

The discrete case. Pieces $i$ with sizes $w_i$; the share of a selection $S$ is $\sum_{i\in S} w_i \;/\; \sum_i w_i$. Instances: softmax/attention weights in neural networks (tesseractions computed trillions of times daily); the Boltzmann distribution, whose denominator is literally called the partition function; weighted averages; the medieval rule of alligation.
The integral case. With density $\rho$, the share of $[a,b]$ is $\int_a^b\rho \;/\; \int\rho$. When $\rho$ is normalized, the cumulative function $F(x)=\int_{-\infty}^x \rho$ is the “running tesseraction from the left edge,” and its inverse $F^{-1}$ solves the cutting problem — where do I cut so the piece has a prescribed size? That inverse problem alone generates medians, percentiles, and much of statistics.
Non-uniform rulers. The Riemann–Stieltjes integral $\int f\,dg$ integrates against a ruler whose tick spacing varies — variable-sized pieces built into the integral itself.
Higher dimensions. Exactly as the founding question guessed — product measures and multiple integrals; the share of a country's population in a region is $\iint_{\text{region}}\rho \;/\; \iint_{\text{total}}\rho$.
Curved craft. In computer-aided design, Non-Uniform Rational B-Splines (NURBS) subdivide parameter space by a non-uniform knot vector, shrinking pieces where curvature is high and stretching them where the surface is calm — tesseractions in the service of ship hulls and film monsters.
Partitions of unity. Functions $\varphi_i \ge 0$ with $\sum\varphi_i = 1$; each $\varphi_i(x)$ says “how much of piece $i$ lives at point $x$.” These underpin differential geometry, finite element methods, and quantum measurement.
Unequal denominators in numeration. Mixed-radix systems (hours, minutes, seconds; factorial base); decompositions like $1=\tfrac12+\tfrac13+\tfrac16$ — a whole split into unequal, rule-governed pieces.
The radical frontier. Non-additive measures (Sugeno, the Choquet integral) drop additivity itself, so the whole need not equal the sum of its parts — used in decision theory and game theory where pieces cooperate or interfere.

The sanity check. Any design must degenerate to ordinary fractions when the rule is constant and the tiles equal. Measure theory does; so shall we. And one honest caveat: as numbers, tesseraction values are just reals — the new content is never the value, it is the structure (which whole, which rule, which piece) and the logic of how shares combine. But that is equally true of $\tfrac34$: the power of fractions was never in the numbers, but in the algebra of parts and wholes.

The Plan. Even mapped territory rewards independent exploration, and one niche is genuinely underserved: notation that treats the partition as a first-class citizen rather than collapsing to a bare value. We therefore design the system anyway — arithmetic first, in the manner of the AddSubMulDivia series; then the calculus, as selected from IntDiffLogExpia. The organizing drama will be closure: each operation either keeps us home or forces a named extension. This is the childhood of number systems — $\mathbb{N},\mathbb{Z},\mathbb{Q},\mathbb{R}$ — re-run at the level of parts and wholes.

Part I · AddSubMulDivia
The Book of the Four Tesseractions

Chapter 2 · Foundations

Definition 2.1 (Opus). A whole or opus is $\Omega=(I,w)$: a finite index set $I$ with weights $w:I\to\mathbb{R}_{\ge 0}$ (the tesserae), and total $W := w(I) \gt 0$.

Definition 2.2 (Tesseraction). A tesseraction is $\sigma=\tes{A}{\Omega}{w}$ with $A\subseteq I$; its value is $\tval{A}{\Omega}{w} := \dfrac{w(A)}{W}$.

Definition 2.3 (Vermiculation). $\Omega'$ refines $\Omega$ via a surjection $\pi:I'\to I$ with $w(i)=\sum_{i'\in\pi^{-1}(i)} w'(i')$ — every tessera may be split; weight is conserved.

Theorem 2.1 (Van's Theorem of Equivalent Shares). Refinement preserves value: $\tval{\pi^{-1}(A)}{\Omega'}{w'} = \tval{A}{\Omega}{w}$.

Proof. Finite sums may be regrouped — entered into the record as the Axiom of the Tile (see Errata). ∎

Corollary (Equitesseral Cancellation). $\frac{a}{b}=\frac{ka}{kb}$: split each of $b$ equal tesserae into $k$. Rational numbers were always equivalence classes of expressions under refinement — we merely admit it.

Definition 2.4 (Two Combinators). Coproduct $\Omega_1\sqcup\Omega_2$: juxtapose the wholes; totals add. Product $\Omega_1\otimes\Omega_2 := (I\times J,\, w\otimes u)$; totals multiply.

The Continuous Mirror. Replace $(I,w)$ by a bounded domain $\Omega\subset\mathbb{R}^n$ with a non-negative integrable tiling rule $\rho$, and sums by integrals: $$\mu_\rho(A)=\int_A \rho(x)\,dx, \qquad \tval{A}{\Omega}{\rho}=\frac{\int_A\rho\,dx}{\int_\Omega\rho\,dx}.$$ Theorem 2.2 (Equitesseral Check, Mirrored). If $\rho = k$ constant and $\Omega=[0,1]$ is cut into $n$ equal segments, every segment has value $1/n$: standard fractions are the zeroth-order, constant-density special case.

Proof. $\int_{x_{i-1}}^{x_i} k\,dx = k/n$; the denominator is $k$. ∎

Proposition 2.3 (The Bridge). Vermiculation pursued without end is integration: Riemann sums are Theorem 2.1 holding at every finite stage, and the integral is the value they settle on. Part II makes this official.

🐍 Python companion — the tesseraction builder (Defs. 2.1–2.2)

Build an opus, choose a selection, watch the value; then equalize to run the equitesseral check.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

# --- The opus: tesserae with variable weights (Defs. 2.1-2.2) --------------
w = np.array([3.0, 1.0, 4.0, 1.5, 2.5, 1.0])
A = {0, 2, 4}                       # the selection: WHICH tesserae

def value(w, A):
    return w[sorted(A)].sum() / w.sum()     # |T| = w(A)/W

print(f"|T| = {value(w, A):.4f}")

# --- Equitesseral check: equal weights recover the classical fraction -----
w_eq = np.full_like(w, w.mean())
print(f"equalized: each tessera has value {value(w_eq, {0}):.4f} = 1/{len(w)}")

fig, ax = plt.subplots(figsize=(8, 3))
colors = [ACCENT if i in A else 'lightsteelblue' for i in range(len(w))]
ax.bar(range(len(w)), w, color=colors, edgecolor='k')
for i, wi in enumerate(w):
    ax.text(i, wi + 0.05, f"{wi:g}", ha='center')
ax.set(xlabel='tessera index i', ylabel='weight $w_i$',
       title=f"selection $A$ (red):  $|T| = w(A)/W = {value(w, A):.4f}$")
plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

🐍 Python companion — vermiculation preserves value (Thm 2.1)

Split every tessera into k; the machine is honest: the value never moves.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import animation

w0 = np.array([3.0, 1.0, 4.0, 1.5, 2.5, 1.0])
A0 = np.array([0, 2, 4])

fig, ax = plt.subplots(figsize=(8, 3))
def draw(k):
    ax.clear()
    wk = np.repeat(w0, k) / k                    # split each tessera into k
    sel = np.isin(np.arange(len(wk)) // k, A0)   # lift of the selection
    v = wk[sel].sum() / wk.sum()
    ax.bar(np.arange(len(wk)), wk, width=1.0,
           color=np.where(sel, ACCENT, 'lightsteelblue'))
    ax.set(ylim=(0, w0.max() * 1.1),
           title=f"split factor k = {k}:  $|T| = {v:.6f}$  (Thm 2.1: invariant)")

ani = animation.FuncAnimation(fig, draw, frames=[1, 2, 3, 4, 6, 8, 12], interval=900)
# Colab:  from IPython.display import HTML; HTML(ani.to_jshtml())
# Web:    ani.save('py02_vermiculator.gif', writer='pillow', fps=2)
plt.show()

⬇ notebook · paste the code into colab.new → Run

🐍 Python companion — the continuous mirror (Thm 2.2)

A fixed tiling rule (the boundary layer); the right handle sweeps; the share is the running tesseraction's difference. (Rule-switching belongs to py14's gauge probe — one handle per companion.)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import animation

rho = lambda x: 1.0 + 9.0*np.exp(-x/0.05)      # the boundary-layer rule
x = np.linspace(0, 1, 4001)
mass = np.cumsum(rho(x)); mass /= mass[-1]     # the running tesseraction F(x)

a = 0.15
fig, ax = plt.subplots(figsize=(8, 3.5))
def draw(t):
    ax.clear()
    b = 0.15 + 0.75 * t
    ax.plot(x, rho(x), color=NAVY, lw=2)
    m = (x >= a) & (x <= b)
    ax.fill_between(x[m], rho(x)[m], color=ACCENT, alpha=0.5)
    v = np.interp(b, x, mass) - np.interp(a, x, mass)
    ax.set(ylim=(0, 11), title=f"share of $[a,b]$ = {v:.4f}   (the running tesseraction)")

ani = animation.FuncAnimation(fig, draw, frames=np.linspace(0, 1, 60), interval=60)
# Colab:  from IPython.display import HTML; HTML(ani.to_jshtml())
# Web:    ani.save('py03_mirror.gif', writer='pillow', fps=15)
plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 3 · Addition

Definition 3.1 (Internal Sum). Same opus, disjoint parts: $\tes{A}{\Omega}{w}+\tes{B}{\Omega}{w} := \tes{A\uplus B}{\Omega}{w}$.

Theorem 3.1. The internal sum is well-defined, commutative, associative, has identity $\tes{\emptyset}{\Omega}{w}=0$, and is closed with value $\le 1$.

Theorem 3.2 (Complement). $\tval{A}{\Omega}{w}+\tval{A^c}{\Omega}{w}=1$.

Theorem 3.3 (Inclusion–Exclusion). For overlapping parts, writing $v(\cdot)$ for value within one opus, $$v(A)+v(B)=v(A\cup B)+v(A\cap B).$$ The overlap is counted twice and must be paid back once.

Proof. $w(A)+w(B)$ counts each tessera of $A\cap B$ twice. ∎

Theorem 3.4 (Common Refinement = Cross-Multiplication). In the product opus $\Omega_1\otimes\Omega_2$, the lifts $\bar A = A\times J$ and $\bar B = I\times B$ satisfy $|\bar A|=|\sigma_1|$, $|\bar B|=|\sigma_2|$, $|\bar A\cap \bar B|=|\sigma_1||\sigma_2|$. Hence any sum of two tesseractions is representable in one opus; in the equitesseral case this reads $$\frac{a}{b}+\frac{c}{d}=\frac{ad+bc}{bd}.$$

Proof. $w(A)\,u(J)\,/\,w(I)\,u(J)$. ∎

Theorem 3.5 (The Two Sums — Pooling). Juxtaposing the opuses themselves gives a second honest sum, $\sigma_1\oplus\sigma_2 := \tes{A\sqcup B}{\Omega_1\sqcup\Omega_2}{w\sqcup u}$, with value $$\frac{w(A)+u(B)}{W+U} \;=\; \lambda\,|\sigma_1|+(1-\lambda)\,|\sigma_2|,\qquad \lambda=\tfrac{W}{W+U}.$$ Pooling is a weighted average (the mediant; the “freshman sum” $\frac{a}{b}\oplus\frac{c}{d}=\frac{a+c}{b+d}$), and always lies between its inputs. Refined sum answers “how much in total?”; pooled sum answers “how much across the combined enterprise?” Batting averages pool; they do not add.

The Continuous Mirror. For tesseractions over $(\Omega_A,f)$ and $(\Omega_B,g)$, juxtapose the domains and piece the rules: $$\tval{D_A\sqcup D_B}{\Omega_A\sqcup\Omega_B}{f\sqcup g} =\frac{\mu_f(D_A)+\mu_g(D_B)}{\mu_f(\Omega_A)+\mu_g(\Omega_B)} =\lambda\,|T_A|+(1-\lambda)\,|T_B|,\qquad \lambda=\tfrac{\mu_f(\Omega_A)}{\mu_f(\Omega_A)+\mu_g(\Omega_B)}.$$ The discrete and continuous pooling formulas are the same formula — the mirror's first perfect reflection. This is the exact law for merging two non-homogeneous fluid tanks.

Theorem 3.6 (Simpson). Pooling can reverse componentwise order.

Example. $\frac{9}{10} \gt \frac{89}{100}$ and $\frac{10}{100}\gt\frac{0}{10}$, yet $\frac{9}{10}\oplus\frac{10}{100}=\frac{19}{110} \lt \frac{89}{110}=\frac{89}{100}\oplus\frac{0}{10}$. The weights $\lambda$ differ between pools; that is the whole paradox. ∎

🐍 Python companion — inclusion–exclusion (Thm 3.3)

Random opus, two overlapping selections; the overlap is paid back exactly, every time.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

rng = np.random.default_rng(7)
w = rng.uniform(0.5, 3.0, size=8); W = w.sum()
v = lambda S: w[sorted(S)].sum() / W if S else 0.0

A, B = {0, 1, 2, 3}, {2, 3, 4, 5}
lhs = v(A) + v(B) - v(A & B)
rhs = v(A | B)
print(f"v(A)+v(B)-v(overlap) = {lhs:.6f}   v(union) = {rhs:.6f}   "
      f"paid back exactly: {np.isclose(lhs, rhs)}  ✓ (Thm 3.3)")

fig, ax = plt.subplots(figsize=(8, 3))
group = ['both' if i in A and i in B else 'A only' if i in A
         else 'B only' if i in B else 'neither' for i in range(8)]
cmap = {'both': GOLD, 'A only': ACCENT, 'B only': NAVY, 'neither': 'lightgrey'}
ax.bar(range(8), w, color=[cmap[g] for g in group], edgecolor='k')
ax.set(xlabel='tessera', ylabel='weight',
       title='gold tesserae are counted twice by v(A)+v(B) — paid back once')
plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

🐍 Python companion — pooling and Simpson's reversal (Thms. 3.5–3.6)

The mediant is a λ-weighted average; load the Simpson preset and watch pooled defeat snatch victory.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

def pool(a, b, c, d):          # the mediant: (a/b) ⊕ (c/d)
    return (a + c) / (b + d)

# Simpson preset from Thm 3.6: team X wins BOTH seasons, loses the pool
X1, X2 = (9, 10), (89, 100)    # (hits, at-bats)
Y1, Y2 = (10, 100), (0, 10)
print(f"season 1: {X1[0]/X1[1]:.3f} > {Y1[0]/Y1[1]:.3f}   "
      f"season 2: {X2[0]/X2[1]:.3f} > {Y2[0]/Y2[1]:.3f}")
print(f"pooled:   {pool(*X1, *X2):.3f} < {pool(*Y1, *Y2):.3f}   — reversal! ✓ (Thm 3.6)")

# Why: pooling is a weighted average whose weight λ = W/(W+U) differs per team
s1x, s2x = X1[0]/X1[1], X2[0]/X2[1]
s1y, s2y = Y1[0]/Y1[1], Y2[0]/Y2[1]
fig, ax = plt.subplots(figsize=(7, 4))
ax.plot([s1x, s2x], [1, 2], 'o-', color=ACCENT, label='team X seasons')
ax.plot([s1y, s2y], [1, 2], 's-', color=NAVY, label='team Y seasons')
ax.plot([pool(*X1, *X2), pool(*Y1, *Y2)], [1, 2], 'd', color=GOLD, ms=12, label='pooled')
for s, y, t in [(s1x,1,'X1'),(s2x,2,'X2'),(s1y,1,'Y1'),(s2y,2,'Y2')]:
    ax.annotate(t, (s, y), textcoords='offset points', xytext=(8, 0))
ax.set(yticks=[1, 2], yticklabels=['pool 1', 'pool 2'], xlabel='batting average',
       title='Simpson: each pooled point is a λ-weighted average of its two seasons')
ax.legend(); plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 4 · Subtraction

Definition 4.1 (Relative Complement). For $B\subseteq A$ in one opus: $\tes{A}{\Omega}{w}-\tes{B}{\Omega}{w}:=\tes{A\setminus B}{\Omega}{w}$.

Theorem 4.1 (Partiality). The difference is a proper tesseraction iff the subtrahend's part nests inside the minuend's. Otherwise values obey $v(A)-v(B)=v(A\setminus B)-v(B\setminus A)$ — a difference of two disjoint tesseractions.

Theorem 4.2 (Closure Forces Signed Tesserae; the Jordan Decomposition). Admitting $\mathbb{R}$-valued weights closes the system under subtraction, and every signed opus splits uniquely as positive part minus negative part on disjoint supports.

Proof. Set $w^+=\max(w,0)$, $w^-=\max(-w,0)$; then $w=w^+-w^-$ and the supports are disjoint by construction. ∎

This is signed measure theory's fundamental theorem, appearing here in its native habitat: the arithmetic of taking away.

The Continuous Mirror. For $D_B\subseteq D_A\subseteq\Omega$: $\tval{D_A\setminus D_B}{\Omega}{\rho}=|T_A|-|T_B|$; and the absolute complement reads $\tval{\Omega\setminus D}{\Omega}{\rho}=1-|T|$. Nothing new — which is the point. Subtraction is the mirror's calmest room.

🐍 Python companion — signed tesserae, split uniquely (Thm 4.2)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

w = np.array([3.0, -2.0, 4.0, -1.0, -2.5, 1.5])
wp, wm = np.maximum(w, 0), np.maximum(-w, 0)

assert np.allclose(w, wp - wm)
assert not np.any((wp > 0) & (wm > 0))      # disjoint supports: uniqueness
print(f"w  = {w}\nw+ = {wp}\nw- = {wm}")
print(f"net = {wp.sum() - wm.sum():g} = w⁺ {wp.sum():g} − w⁻ {wm.sum():g}  ✓ (Thm 4.2)")

fig, ax = plt.subplots(figsize=(8, 3))
ax.bar(np.arange(6) - 0.2, wp, width=0.4, color='seagreen', label='$w^+$')
ax.bar(np.arange(6) + 0.2, -wm, width=0.4, color=ACCENT, label='$-w^-$')
ax.axhline(0, color='k', lw=0.8)
ax.set(xlabel='tessera', title='Jordan decomposition: $w = w^+ - w^-$ on disjoint supports')
ax.legend(); plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 5 · Multiplication

Definition 5.1 (Product Tesseraction). $\tes{A}{\Omega_1}{w}\cdot\tes{B}{\Omega_2}{u} := \tes{A\times B}{\Omega_1\otimes\Omega_2}{w\otimes u}$.

Theorem 5.1 (Value Multiplies). $|\sigma_1\sigma_2|=|\sigma_1|\,|\sigma_2|$.

Proof. $w(A)\,u(B)\,/\,WU$. ∎

Reading: “a share of a share” — two-stage sampling is the Cartesian product.

Theorem 5.2. Multiplication is commutative and associative up to canonical isomorphism of opuses; the identity is the trivial one-tessera opus; it commutes with vermiculation and distributes over disjoint addition, since $(A\uplus B)\times C=(A\times C)\uplus(B\times C)$.

Theorem 5.4 (Telescoping — the Chain Rule). If $A\subseteq B\subseteq C$: $\;\tes{A}{B}{w}\cdot\tes{B}{C}{w}=\tes{A}{C}{w}$.

Proof. $\frac{w(A)}{w(B)}\cdot\frac{w(B)}{w(C)}=\frac{w(A)}{w(C)}$. ∎

Tesseractions compose multiplicatively along chains of sub-opuses. (The cognoscenti will recognize conditional probability's chain rule; per house rules, we pretend not to know this until the sequel.)

The Continuous Mirror — Fubini. With product density $(f\otimes g)(x,y)=f(x)g(y)$ on $\Omega_A\times\Omega_B$, $$\tval{D_A\times D_B}{\Omega_A\times\Omega_B}{f\otimes g} =\frac{\int_{D_A}\!f\,dx\int_{D_B}\!g\,dy}{\int_{\Omega_A}\!f\,dx\int_{\Omega_B}\!g\,dy} =|T_A|\cdot|T_B|,$$ which is Fubini's theorem — i.e., Theorem 5.1 seen in the mirror. The discrete and continuous product laws are one law. Theorem 5.5 (Conditional Restriction; the Bayes–Radon Chain). Restricting the opus to $D_B$: $\tval{D_A\cap D_B}{D_B}{\rho}\cdot\tval{D_B}{\Omega}{\rho}=\tval{D_A\cap D_B}{\Omega}{\rho}$ — the mirror image of telescoping.

🐍 Python companion — the product mosaic (Thm 5.1)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.patches import Rectangle

w = np.array([3.0, 1.0, 4.0, 1.5]);  A = [0, 2]     # columns: opus 1
u = np.array([2.0, 1.0, 3.0]);       B = [1, 2]     # rows:    opus 2

P = np.outer(u, w)                      # the product opus w ⊗ u, shown as a mosaic
block = P[np.ix_(B, A)].sum() / P.sum()
prod  = (w[A].sum() / w.sum()) * (u[B].sum() / u.sum())
print(f"block share = {block:.6f}   |T1|·|T2| = {prod:.6f}   "
      f"equal: {np.isclose(block, prod)}  ✓ (Thm 5.1)")

fig, ax = plt.subplots(figsize=(6, 4))
ax.imshow(P, cmap='Blues')
for i in range(len(u)):
    for j in range(len(w)):
        ax.text(j, i, f"{P[i, j]:g}", ha='center', va='center')
ax.add_patch(Rectangle((min(A)-0.5, min(B)-0.5), len(A), len(B),
                       fill=False, edgecolor=ACCENT, lw=2.5))
ax.set_title(f"$A \\times B$ highlighted: share {block:.4f} = $|T_1||T_2|$ (Thm 5.1)")
plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

🐍 Python companion — telescoping (Thm 5.4)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

wA, wB, wC = 2.0, 5.0, 12.0        # nested parts A ⊆ B ⊆ C of one opus
lhs = (wA / wB) * (wB / wC)        # the middle weight cancels
rhs = wA / wC
print(f"(wA/wB)(wB/wC) = {lhs:.6f}   wA/wC = {rhs:.6f}   "
      f"equal: {np.isclose(lhs, rhs)}  ✓ telescoped (Thm 5.4)")

fig, ax = plt.subplots(figsize=(7, 2.5))
for y, (val, name, col) in enumerate([(wC, 'C (opus)', 'lightsteelblue'),
                                      (wB, 'B', NAVY),
                                      (wA, 'A', ACCENT)]):
    ax.barh(y, val, color=col, edgecolor='k')
    ax.text(val + 0.15, y, f"{name}: {val:g}", va='center')
ax.set(yticks=[], xlabel='weight',
       title='chains of sub-opuses multiply: the middle link cancels')
plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 6 · Division

Division forces the syntax to grow. A generalized tesseraction $\tes{A}{B}{w}=w(A)/w(B)$ requires no containment; it is proper iff $A\subseteq B$, improper otherwise (selections may be multisets — copies count).

Theorem 6.1 (The Reciprocal Swap — Flagship of the Book). Define the reciprocal by swapping part and whole: $$\left(\tes{A}{B}{w}\right)^{-1}:=\tes{B}{A}{w}.$$ Reciprocation is an involution, and $\sigma\cdot\sigma^{-1}=1$ by Telescoping with $A=B$.

Proof. $\frac{w(A)}{w(B)}\cdot\frac{w(B)}{w(A)}=1$; applied twice, the swap restores the original. ∎

Theorem 6.2 (Division). $\sigma_1/\sigma_2 = \tes{A\times J}{I\times B}{w\otimes u}$ in the product weighting, with value $\dfrac{w(A)\,U}{W\,u(B)}$. The result is typically improper — division dilates, and $[0,1]$ cannot hold it.

Theorem 6.3 (Quotitive Meaning). $\sigma_1/\sigma_2$ counts how many copies of the second part fit inside the first: division is measurement.

Theorem 6.4 (Euclidean Division — Mixed Shares). Every improper tesseraction decomposes uniquely as $$\frac{w(A)}{w(B)} \;=\; q+\frac{w(R)}{w(B)},\qquad q=\lfloor w(A)/w(B)\rfloor,\quad w(R)=w(A)-q\,w(B),$$ the division algorithm on weights; mixed numbers are tesseractions wearing their wholes openly.

The Continuous Mirror. The mirror shows the same face, and here the founders' two schools arrived independently at identical constructions: for $\mu_\rho(D)\gt 0$, $$\left(\tes{D}{\Omega}{\rho}\right)^{-1}=\tes{\Omega}{D}{\rho},\qquad \left|\tes{\Omega}{D}{\rho}\right|=\frac{\mu_\rho(\Omega)}{\mu_\rho(D)}=|T|^{-1},$$ the Universe Swap. Division is then tensor multiplication by the swapped reciprocal, $\tes{D_A\times\Omega_B}{\Omega_A\times D_B}{f\otimes g}$, value $|T_A|/|T_B|$. Part and whole change places; the geometry does the algebra.

🐍 Python companion — quotitive division and the reciprocal swap (Thms. 6.1, 6.4)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import animation

wA, wB = 11.0, 4.0
q = int(wA // wB); r = wA - q * wB    # Euclidean division on weights (Thm 6.4)
print(f"{wA}/{wB} = {q} + {r}/{wB}    (quotient {q}, remainder {r})")
print(f"reciprocal swap (Thm 6.1): {wB}/{wA} = {wB/wA:.4f} = 1/{wA/wB:.4f}")

fig, ax = plt.subplots(figsize=(8, 2.5))
def draw(t):
    ax.clear()
    filled = min(t * (q + 1), q + r / wB)   # stack whole copies of B, then the remainder
    whole = int(filled)
    for i in range(whole):
        ax.barh(0, wB, left=i * wB, color=NAVY if i % 2 else ACCENT, edgecolor='k')
    frac = filled - whole
    if frac > 1e-9:
        ax.barh(0, wB * frac, left=whole * wB, color=GOLD, edgecolor='k')
    ax.barh(0, wA, fill=False, edgecolor='k', lw=2)
    ax.set(xlim=(-0.2, wA + 0.5), yticks=[],
           title=f"how many B's fit in A?  {filled:.2f}  →  {q} + {r:g}/{wB:g}  (Thm 6.4)")

ani = animation.FuncAnimation(fig, draw, frames=np.linspace(0, 1, 60), interval=60)
# Colab:  from IPython.display import HTML; HTML(ani.to_jshtml())
# Web:    ani.save('py09_quotitive.gif', writer='pillow', fps=15)
plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 7 · Powers — or, The Ascent of the Tesseract

“A line, a square, a cube — and then the stones run out of directions before the tiles run out of ascent.”— after C. H. Hinton, who named the fourth rung

Definition 7.1 (The Ascent). For $\sigma=\tes{A}{\Omega}{w}$ and $n\ge 1$: $$\sigma^n:=\tes{A^n}{\Omega^{\otimes n}}{w^{\otimes n}},\qquad A^n:=A\times\cdots\times A\subseteq I^n,$$ with $\sigma^1:=\sigma$ and $\sigma^0:=\mathbf 1$ (the empty product of opuses is the one-tessera opus, fully selected — the empty ascent leaves you standing on the ground floor, which is 1).

Theorem 7.1 (Value of an Ascent). $|\sigma^n|=|\sigma|^n$.

Proof. Induction on Theorem 5.1. ∎

Lemma 7.2 (Monotonicity of Ascent). If $0\le|\sigma| \lt |\tau|$ then $|\sigma|^n \lt |\tau|^n$.

Proof. $b^n-a^n=(b-a)(b^{n-1}+b^{n-2}a+\cdots+a^{n-1})$, a product of positives. Hence ascents are faithful: distinct values never share a rung, and roots, when they exist, are unique in value. ∎

Theorem 7.3 (The Index Laws). Up to canonical isomorphism: $$\sigma^m\cdot\sigma^n\cong\sigma^{m+n},\qquad (\sigma^m)^n\cong\sigma^{mn},\qquad (\sigma\tau)^n\cong\sigma^n\tau^n.$$

Proof. Concatenation $I^m\times I^n\cong I^{m+n}$; reindexing $(I^m)^n\cong I^{mn}$; the shuffle that sorts mixed coordinates. On values these become the familiar laws — now theorems of tilings rather than axioms of symbols. ∎

Theorem 7.4 (The Bi-infinite Ladder). With $\sigma^{-n}:=(\sigma^{-1})^n$: for nondegenerate $\sigma$ (value not $0$ or $1$) the values $|\sigma|^z$, $z\in\mathbb{Z}$, are all distinct — positive rungs proper, rung zero the unity, negative rungs improper.

Proof. If $|\sigma|^a=|\sigma|^b$ then $|\sigma|^{a-b}=1$, forcing $|\sigma|=1$ by Lemma 7.2. ∎

Theorem 7.5 (Trichotomy of Ascent). Let $x=|\sigma|$. (i) $x \lt 1$: the ascent strictly descends and $x^n\to 0$ — it undercuts every bound. (ii) $x=1$: fixed at every rung. (iii) $x \gt 1$ (improper): strictly grows beyond every bound.

Proof. (ii) is immediate. For (i): by Chapter 6, $1/x=1+h$, $h \gt 0$. Bernoulli's Inequality: $(1+h)^n\ge 1+nh$, by induction: $(1+nh)(1+h)=1+(n{+}1)h+nh^2\ge 1+(n{+}1)h$. The weight world is Archimedean — given any bound, finitely many tiles suffice to pass it — so $(1+h)^n$ exceeds all bounds and $x^n$ undercuts all bounds. (iii) is (i) read upward. ∎

Theorem 7.6 (The Tesseract Theorem). The fourth power of any tesseraction is a tesseract of tesserae: $\sigma^4=\tes{A^4}{\Omega^{\otimes 4}}{w^{\otimes 4}}$ is a four-dimensional lattice of $|I|^4$ tiles, its selected region a four-dimensional block of $|A|^4$ tiles — and the ascent $\sigma,\sigma^2,\sigma^3,\sigma^4$ realizes in turn the line, the square, the cube, and the tesseract.

Proof. By definition of the ascent; the theorem is that the climb is real. Hinton named the four-cube in 1888; the act that builds it was named in this book, and the names agree. ∎

Corollary 7.7 (The Vanishing Tesseract). If $\sigma$ is proper, the ascents vanish: $|\sigma|^n\to 0$. The selected corner of the tesseract shrinks rung by rung; the tesserae dwindle to dust; the tesseract, iterated within itself, vanishes.

Proof. Trichotomy (i). ∎

§7.5 · The Chamber Decomposition

The selected block $A^n$ is but one corner of the $n$-cube. For each $S\subseteq\{1,\dots,n\}$, the chamber $C_S$ takes its $k$-th coordinate from $A$ when $k\in S$, from $A^c$ otherwise.

Theorem 7.8 (The Tesseractic Binomial Theorem). The chambers partition $\Omega^{\otimes n}$; chamber $C_S$ carries value $|\sigma|^{|S|}\,|\sigma^c|^{\,n-|S|}$; chambers with $|S|=k$ number $\binom{n}{k}$; whence $$1=\big(|\sigma|+|\sigma^c|\big)^n=\sum_{k=0}^{n}\binom{n}{k}\,|\sigma|^k\,|\sigma^c|^{\,n-k}.$$

Proof. Disjointness of chambers, Theorem 5.1 chamberwise, and the count of $k$-subsets. ∎

Corollary 7.9 (Pascal's Rule). $\binom{n}{k}=\binom{n-1}{k-1}+\binom{n-1}{k}$ — a chamber's last coordinate either lies in $A$ or does not.

The Freshman's Dream $(\sigma\oplus\tau)^n=\sigma^n\oplus\tau^n$ fails precisely by the mixed chambers — Pascal enumerates the dream's error term. (The cognoscenti will recognize the binomial distribution standing in the doorway; per house rules, we pretend not to know her until the sequel.)

§7.6 · The Descent: Roots

Theorem 7.10 (Hippasus; the First Incommensurable). No equitesseral tesseraction has square of value 2.

Proof. Suppose $p^2=2q^2$ in lowest terms. Then $p$ is even, $p=2r$, whence $q^2=2r^2$ and $q$ is even — contradiction. ∎

Theorem 7.11 (Closure under Roots). In the full (real-weighted) system every value $x\in[0,1]$ has a unique $n$-th root; two tesserae suffice: weights $\big(x^{1/n},\,1-x^{1/n}\big)$. Uniqueness is Lemma 7.2. Read with Theorem 7.10: the equitesseral world refuses roots; the tesseractic world grants them. Closure does not manufacture new tesserae — it manufactures new values. The irrationals enter not as tesserae but as tesseraction-values.

Theorem 7.12 (Rational Ascent, Well-Defined). $\sigma^{m/n}:=(\sigma^m)^{1/n}$ depends only on the value of $m/n$, and equals $(\sigma^{1/n})^m$.

Proof. Both are the unique value $y$ with $y^n=|\sigma|^m$, by Lemma 7.2 and the Index Laws. ∎

Theorem 7.13 (The Mean Proportional). For any two tesseractions there is a unique-in-value $\gamma$ with $|\gamma|^2=|\sigma_1||\sigma_2|$, and $|\gamma|\le\tfrac12(|\sigma_1|+|\sigma_2|)$.

Proof. Existence by 7.11; the inequality from $0\le(a-b)^2=(a+b)^2-4ab$. This is Euclid VI.13 in tesserae — roots were born as mean proportionals long before they were born as symbols. ∎

Theorem 7.14 (The Ascent Sum, Finite). For $x=|\sigma|\ne 1$: $\;1+x+x^2+\cdots+x^n=\dfrac{1-x^{\,n+1}}{1-x}$.

Proof. $(1-x)(1+x+\cdots+x^n)=1-x^{n+1}$: each rung cancels its successor — Chapter 4 subtraction, iterated. ∎

A boundary stone. For proper $\sigma$, the Vanishing Tesseract erases the final rung $x^{n+1}$ as the ascent lengthens, and the sums approach $\frac{1}{1-x}$. But approach is a word Part I has not licensed. Licenses are issued in Part II.

🐍 Python companion — the rotating tesseract of tesserae, and its vanishing (Thms. 7.6–7.7)

Vertices $\{0,1,2\}^4$, selected corner $A^4$ with $A=\{1,2\}$ ($x = 2/3$), rotated in the $xw$-plane and projected to the page; the decay plot runs the Vanishing Tesseract in parallel. (Rungs 1–3: replace $V$ by $\{0,1,2\}^n$ — line, square, cube.)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import animation

V = np.array([[i, j, k, l] for i in (0, 1, 2) for j in (0, 1, 2)
                            for k in (0, 1, 2) for l in (0, 1, 2)], float) - 1.0
sel = np.all(np.abs(V) > 0.5, axis=1)                 # the A^4 sub-tesseract (16 vertices)
E = [(a, b) for a in range(81) for b in range(a + 1, 81)
     if np.sum(np.abs(V[a] - V[b])) == 1.0]            # unit edges of the 4-grid

def project(P, th):
    c, s = np.cos(th), np.sin(th)
    R = np.eye(4); R[0, 0] = R[3, 3] = c; R[0, 3] = -s; R[3, 0] = s   # xw rotation
    Q = P @ R.T
    d = 2.5
    return Q[:, :3] / (d - Q[:, 3])[:, None] * d       # 4D→3D perspective, then drop z

fig = plt.figure(figsize=(10, 4))
ax1, ax2 = fig.add_subplot(121), fig.add_subplot(122)
ns = np.arange(1, 21)
ax2.plot(ns, (2/3)**ns, 'o-', color=ACCENT)
ax2.set(xlabel='rung n', ylabel='$|T|^n$',
        title='Vanishing Tesseract: $(2/3)^n \\to 0$ (Cor. 7.7)')

def draw(th):
    ax1.clear(); ax1.axis('off')
    ax1.set(xlim=(-1.7, 1.7), ylim=(-1.7, 1.7), aspect='equal',
            title='$\\sigma^4$: a tesseract of tesserae (Thm 7.6)')
    P = project(V, th)
    for a, b in E:
        hot = sel[a] and sel[b]
        ax1.plot([P[a, 0], P[b, 0]], [P[a, 1], P[b, 1]],
                 color=ACCENT if hot else 'lightsteelblue', lw=2.0 if hot else 0.7)
    ax1.scatter(P[sel, 0], P[sel, 1], c=ACCENT, s=18, zorder=3)

ani = animation.FuncAnimation(fig, draw, frames=np.linspace(0, np.pi, 120), interval=50)
# Colab:  from IPython.display import HTML; HTML(ani.to_jshtml())
# Web:    ani.save('py10_tesseract.gif', writer='pillow', fps=20)
plt.show()

⬇ notebook · paste the code into colab.new → Run

🐍 Python companion — chambers counted by Pascal, weighted by x (Thm 7.8)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import animation
from math import comb

n = 8
fig, ax = plt.subplots(figsize=(8, 3.5))
def draw(x):
    ax.clear()
    k = np.arange(n + 1)
    counts = np.array([comb(n, int(ki)) for ki in k])
    vals = counts * x**k * (1 - x)**(n - k)
    ax.bar(k - 0.2, counts / counts.max(), width=0.4, color=NAVY,
           label='chamber counts (normalized)')
    ax.bar(k + 0.2, vals, width=0.4, color=ACCENT, label='chamber values')
    ax.set(xlabel='k', title=f"n = {n}, x = {x:.2f}:  Σ values = {vals.sum():.6f}  (Thm 7.8: always 1)")
    ax.legend()

ani = animation.FuncAnimation(fig, draw, frames=np.linspace(0.05, 0.95, 60), interval=80)
# Colab:  from IPython.display import HTML; HTML(ani.to_jshtml())
# Web:    ani.save('py11_chambers.gif', writer='pillow', fps=15)
plt.show()

⬇ notebook · paste the code into colab.new → Run

Part II · Selections from IntDiffLogExpia
The Book of Approach

Chapter 8 · The Axiom of Approach

“That which is in locomotion must arrive at the half-way stage before it arrives at the goal.”— Zeno, by way of Aristotle, presenting the first unlicensed limit

Part I ran on one property alone — the Archimedean, finitely many tiles suffice to pass any bound — and refused the word approach at every door. Three debts stand outstanding: the infinite ascent sum, the gaps in the ladder (irrational rungs), and the Ladder Question itself. One axiom pays all three debts. This is the entire cost of calculus, and we pay it here, once.

Axiom of Approach (Axiom A). Every ascending sequence of tesseraction values bounded beneath a ceiling settles on a value: if $x_1\le x_2\le\cdots\le C$, there is a unique $L$ — the limit — such that for every $\varepsilon \gt 0$, all terms past some stage lie within $\varepsilon$ of $L$.

Tesseractic restatement. An endless vermiculation whose tesserae shrink without bound settles on a definite content. The opus does not dissolve under infinite subdivision; it arrives.

Homage, in order of appearance: Eudoxus, whose method of exhaustion was this axiom wearing a disguise; Archimedes, who exhausted a circle and called it measurement; Stevin, who wrote real numbers as unending decimals and dared anyone to object.

Theorem 8.1 (Uniqueness). A sequence approaches at most one value.

Proof. Two putative limits lie within $2\varepsilon$ of each other for every $\varepsilon$; the Archimedean property leaves them no room to differ. ∎

Theorem 8.2 (The Four Tesseractions Commute with Approach). If $a_n\to a$ and $b_n\to b$: $$a_n\pm b_n\to a\pm b,\qquad a_nb_n\to ab,\qquad a_n/b_n\to a/b\;\;(b\ne0).$$

Proof. $|a_nb_n-ab|\le|a_n||b_n-b|+|b||a_n-a|$, each summand halved by waiting long enough; the rest is bookkeeping with $\varepsilon/2$. The moral is structural: Part I's algebra survives passage to the limit. ∎

Theorem 8.3 (The Infinite Ascent Sum). For proper $\sigma$ with $x=|\sigma|$: $$1+x+x^2+x^3+\cdots=\frac{1}{1-x}.$$

Proof. By the finite Ascent Sum (7.14) the partial sum is $\frac{1-x^{N+1}}{1-x}$; by the Vanishing Tesseract (7.7), $x^{N+1}\to 0$; by Theorem 8.2 the quotient approaches $\frac{1}{1-x}$. The prophesied theorem becomes the workhorse of the sequel. ∎

Corollary 8.4 (Zeno's Tiling — the Halving Cascade). $$\frac12+\frac14+\frac18+\cdots=1.$$ Tesseractic reading. An opus may be tiled by infinitely many tesserae: take half, then half the remainder, then half of that. Each tessera is the same tesseraction — one-half — of what remains; the tiling is self-similar, every tail a scaled copy of the whole. Nothing is left over: the cascade exhausts the opus exactly. Zeno's runner completes the course, for infinitely many acts of arrival compose one finite arrival. This is the theory's first infinite mosaic — and it is precisely a tesseraction: a rule for subdivision, iterated without end, yielding a definite whole.

§8.4 · Irrational Rungs — the Wrinkle, Ironed

Lemma 8.5 (Ground Continuity). $x^{1/n}\to 1$ as $n\to\infty$ (for $x \gt 0$).

Proof. For $0 \lt x\le 1$ the sequence increases toward a limit $L\le 1$ (Axiom A). If $L \lt 1$ then $x\le L^n$ for all $n$ — but $L^n\to 0$ by the Vanishing Tesseract, forcing $x=0$. So $L=1$; $x \gt 1$ follows by reciprocation. ∎

Theorem 8.6 (The Wrinkle). For $x \gt 0$ and real $r$, define $x^r:=\lim x^{q_n}$ along rationals $q_n\to r$. The limit exists, is independent of the approaching sequence, is strictly monotone in $r$, and the Index Laws extend to all real exponents.

Proof. Two rational approaches to $r$ differ by rationals $d_n\to 0$; monotonicity (7.2, 7.12) traps $x^{d_n}$ between $x^{\pm 1/N}$ once $|d_n| \lt 1/N$, and Ground Continuity collapses these to 1. The Index Laws pass from rational identities (7.3) through Theorem 8.2. The tesseract, iterated $\sqrt 2$-fold within itself, vanishes no less surely. ∎

§8.5 · The Ladder Completed: Napier's Bridge

Theorem 8.7 (Existence of the Logarithm). Let $b \gt 0$, $b\ne 1$. Every $x \gt 0$ stands at exactly one rung: there is a unique real $r$ with $b^r=x$.

Proof. Assume $b \gt 1$. Trichotomy (7.5) supplies rational rungs below $x$ (since $b^{-n}\to 0$) and above it (since $b^n$ outruns all bounds). Take $q_n$ the largest dyadic rational with $b^{q_n}\le x$; then $b^{q_n}$ ascends, bounded by $x$, to a limit $L\le x$ (Axiom A), while $x \lt b^{q_n}b^{1/2^n}\to L$ by Ground Continuity. So $x=L=b^r$ with $r=\lim q_n$. Uniqueness is strict monotonicity. ∎

Definition 8.1 (Logarithm). That rung is $\log_b x$ — the height at which $x$ stands on the $b$-ladder. For tesseractions: $\log_\beta\tau := \log_{|\beta|}|\tau|$ — how many $\beta$'s make a $\tau$.

Theorem 8.8 (Napier's Bridge). $$\log_b(xy)=\log_b x+\log_b y,\qquad \log_b(x^r)=r\log_b x,\qquad \log_b x=\frac{\log_c x}{\log_c b}.$$

Proof. Let $b^r=x$, $b^s=y$. Then $xy=b^{r+s}$ by the extended Index Laws — the rung of the product is the sum of the rungs. The Four Tesseractions fold into two: multiplication of tesseractions becomes addition of heights; division, subtraction; powers, multiplication. ∎

“Seeing there is nothing so troublesome to mathematical practice than the multiplications, divisions, and extractions of roots of great numbers… I began to consider by what certain and ready art I might remove those hindrances.”— Napier, 1614

🐍 Python companion — the halving cascade, self-similar (Cor. 8.4)

Each tessera is the same tesseraction — one-half — of what remains; the cascade exhausts the opus exactly.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import animation

fig, ax = plt.subplots(figsize=(9, 2.5))
def draw(k):
    ax.clear()
    x0 = 0.0
    for i in range(k):
        seg = 0.5**(i + 1)                      # tessera i: half of what remains
        ax.barh(0, seg, left=x0, color=mpl.cm.viridis(i / 12), edgecolor='k', lw=0.5)
        x0 += seg
    ax.barh(0, 1.0, fill=False, edgecolor='k', lw=1.5)
    S = 1 - 0.5**k
    ax.set(xlim=(0, 1), yticks=[],
           title=f"k = {k}:  $S_k = 1 - 2^{{-k}}$ = {S:.8f}   remainder $2^{{-k}}$ = {0.5**k:.2e}")

ani = animation.FuncAnimation(fig, draw, frames=range(0, 13), interval=700)
# Colab:  from IPython.display import HTML; HTML(ani.to_jshtml())
# Web:    ani.save('py12_zeno.gif', writer='pillow', fps=2)
plt.show()

# The limit, paid by the Vanishing Tesseract (7.7) via Thm 8.3:
print(f"1/2 + 1/4 + 1/8 + ... = {sum(0.5**k for k in range(1, 60)):.15f} → 1  ✓ (Cor. 8.4)")

⬇ notebook · paste the code into colab.new → Run

🐍 Python companion — the ladder and Napier's Bridge (Thms. 8.7–8.8)

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

b = 2.0
r = np.linspace(-3, 3, 400)
x, y = 5.0, 3.0
logb = lambda v: np.log(v) / np.log(b)

lx, ly, lxy = logb(x), logb(y), logb(x * y)
print(f"log_{b:g} {x:g} + log_{b:g} {y:g} = {lx:.6f} + {ly:.6f} = {lx+ly:.6f}")
print(f"log_{b:g} ({x:g}·{y:g})          = {lxy:.6f}   "
      f"equal: {np.isclose(lx+ly, lxy)}  ✓ (Thm 8.8)")

fig, ax = plt.subplots(figsize=(8, 4))
ax.plot(r, b**r, color=NAVY, lw=2, label=f"ladder ${b:g}^r$")
ax.axhline(1, color='grey', lw=0.8)
for k in range(-3, 4):
    ax.plot(k, b**k, 'o', color=GOLD, ms=5)
ax.plot(lx, x, 'o', color=ACCENT, ms=9)
ax.annotate(f"x = {x:g} stands at rung {lx:.3f}", (lx, x),
            textcoords='offset points', xytext=(10, -14))
ax.set(xlabel='rung r', ylabel='value', title='every value stands at exactly one rung (Thm 8.7)')
ax.legend(); plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 9 · The Natural Rung

“…the number 2.718281828459…, which we shall hereafter denote by the letter e.” — after Euler, Introductio in Analysin Infinitorum, who insisted the letter stood for no one

Theorem 9.1 (Steepness is Proportional to Altitude). $$\frac{b^{x+h}-b^x}{h}=b^x\cdot\frac{b^h-1}{h}.$$ Every ladder ascends at a rate proportional to its own height; the constant depends only on the base.

Proof. $b^{x+h}=b^xb^h$ (Index Laws, extended 8.6); factor. ∎

Definition 9.1 (Gauge). $\gamma(b):=\displaystyle\lim_{h\to0}\frac{b^h-1}{h}$ — the ladder's steepness at ground level. The question of the chapter: is there a base with gauge exactly 1 — a ladder whose steepness at every height equals its height?

§9.2 · The Banker's Tiling

Nothing prevents the tesseraction itself from re-tiling at every rung. Let an opus grow by the tesseraction $\tfrac1n$ at each of $n$ steps, growth reinvested — the improper tesseraction $1+\tfrac1n$, iterated $n$ times: $$a_n:=\left(1+\tfrac1n\right)^n.$$ Homage: Jacob Bernoulli, 1683, who met this sequence counting compound interest — the banker's tiling, compounded ever finer.

Theorem 9.2 (Existence of $e$). $a_n$ strictly increases and is bounded above by $3$; by the Axiom of Approach, $e:=\lim a_n$ exists, with $2.5\le e \lt 3$.

Proof. By the Tesseractic Binomial (7.8): $$a_n=\sum_{k=0}^{n}\frac{1}{k!}\prod_{j=0}^{k-1}\left(1-\tfrac{j}{n}\right).$$ Each factor increases with $n$, and the sum gains a term — hence $a_n$ increases. Each product is $\le 1$, and $k!\ge 2^{k-1}$, so $$a_n\le\sum_{k=0}^{n}\frac{1}{k!}\le 1+\sum_{k\ge1}2^{-(k-1)}=3,$$ the bound paid for by Zeno's Halving Cascade (8.4). Below: $a_n\ge 1+1+\tfrac12(1-\tfrac1n)\to\tfrac52$. ∎

Theorem 9.3 (The Compounded Ladder). For every real $x$, $E(x):=\lim_n\left(1+\tfrac{x}{n}\right)^n$ exists, and $$E(x)E(y)=E(x+y),\qquad E(1)=e,\qquad E\text{ strictly increasing};\qquad\text{hence }E(x)=e^x.$$

Proof sketch. Existence and monotonicity as in 9.2. For $x,y\ge0$: $$\left(1+\tfrac{x}{n}\right)\left(1+\tfrac{y}{n}\right) =\left(1+\tfrac{x+y}{n}\right)\left(1+c_n\right),\qquad c_n=\tfrac{xy}{n(n+x+y)}\to0,$$ and Bernoulli (7.5) traps $1\le(1+c_n)^n\le\tfrac{1}{1-nc_n}\to1$. For negative arguments, $\left(1-\tfrac{x^2}{n^2}\right)^n$ is trapped between $1$ and $1-\tfrac{x^2}{n}$, so $E(-x)E(x)=1$. Then $E(q)=e^q$ for rational $q$, and monotonicity with Theorem 8.6 pins $E$ to the ladder everywhere. Reading: $e^x$ is an opus grown by total tesseraction $x$, compounded in ever-finer vermiculation. The ladder is the limit of all bankers. ∎

Theorem 9.4 (The Master Inequality). For all real $x$: $1+x\le e^x$; and for $x \lt 1$: $e^x\le\dfrac{1}{1-x}$.

Proof. Bernoulli: $(1+\tfrac{x}{n})^n\ge 1+x$ for $n \gt |x|$; pass to the limit. Then $e^{-x}\ge 1-x$ gives the upper bound by reciprocation. ∎

Theorem 9.5 (The Gauge of $e$). $\gamma(e)=1$.

Proof. For $0 \lt h \lt 1$ the Master Inequality traps $$1\le\frac{e^h-1}{h}\le\frac{1}{1-h},$$ and both ends approach 1; $h \lt 0$ by the mirrored squeeze. ∎

Theorem 9.6 (Steepness Equals Height). $\dfrac{e^{x+h}-e^x}{h}\to e^x$ at every height. The natural ladder ascends at exactly the rate it has already ascended — which is why the natural rung governs everything that grows in proportion to itself.

Definition 9.2 (The Natural Rung). $\ln x := \log_e x$. For a tesseraction: $\ln\tau:=\ln|\tau|$. The Sign of Propriety: proper tesseractions stand below ground ($\ln\tau \lt 0$), improper above, unity at rung zero — Part I's boundary, redrawn as ground level.

The Gauge Theorem 9.7. For every base $b \gt 0$: $\gamma(b)=\ln b$.

Proof. $b^h=e^{h\ln b}$ (Theorem 9.3), so $$\frac{b^h-1}{h}=\ln b\cdot\frac{e^{h\ln b}-1}{h\ln b}\longrightarrow\ln b\cdot 1.\;\;∎$$

Corollary 9.8 (Uniqueness). $\gamma$ is strictly increasing with $\gamma(1)=0$; $e$ is the unique gauge-1 base; and since $2 \lt e \lt 3$, $\gamma(2) \lt 1 \lt \gamma(3)$. The steepness constant of any ladder is nothing but its base's natural rung: the two mysteries of the chapter are one mystery.

§9.6 · The Content of the Number

Theorem 9.9 (Euler's Series). $$e=\sum_{k=0}^{\infty}\frac{1}{k!}=1+1+\frac12+\frac16+\frac1{24}+\cdots$$

Proof. From 9.2's expansion: for fixed $K$ and all $n\ge K$, $a_n\ge\sum_{k\le K}\frac1{k!}\prod_{j \lt k}(1-\tfrac jn)$; letting $n\to\infty$ gives $e\ge\sum_{k\le K}\frac1{k!}$; then $K\to\infty$. The reverse bound is $a_n\le\sum_{k\le n}\frac1{k!}$. Squeeze. Each term is a tessera approached from below as the opus vermiculates. ∎

Theorem 9.10 (Fourier). $e$ is irrational.

Proof. Suppose $e=p/q$. Multiply Euler's series by $q!$: the terms through $k=q$ give an integer, leaving remainder $$R=\frac{1}{q+1}+\frac{1}{(q+1)(q+2)}+\cdots \lt \sum_{j\ge1}(q+1)^{-j}=\frac1q\le 1,$$ an infinite ascent sum (8.3). So $0 \lt R \lt 1$ — not an integer, contradiction. (For $q=1$: $2 \lt e \lt 3$.) ∎

A flag planted at the frontier. Hermite, 1873: $e$ satisfies no integer polynomial of any degree — it is transcendental, standing off the algebraic mainland entirely. The quarry is real; the hunt belongs to a later volume.

§9.7 · The Catalogue of Meanings

Presentation	Statement
The banker	$e=\lim_n(1+\tfrac1n)^n$
The mosaic	$e=\sum_k 1/k!$
The gauge	the unique base with $\gamma=1$
The self-similar ladder	steepness equals height, everywhere
The unit rung	$\ln e = 1$

🐍 Python companion — the banker's tiling and the gauge probe (Thms. 9.2, 9.7)

Left: $(1+1/n)^n$ climbing to $e$ between its twin bounds. Right: $(b^h-1)/h \to \ln b$ for any chosen base — at $b = e$ the gauge reads exactly 1.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

b = 2.0                                          # try np.e: gauge snaps to 1
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(11, 3.8))

ns = np.logspace(0, 6, 200)
lo = (1 + 1/ns)**ns                              # banker, increasing (Thm 9.2)
hi = (1 + 1/ns)**(ns + 1)                        # the twin bound, decreasing
ax1.semilogx(ns, lo, color=ACCENT, label=r'$(1+1/n)^n$')
ax1.semilogx(ns, hi, color=NAVY, ls='--', label=r'$(1+1/n)^{n+1}$')
ax1.axhline(np.e, color=GOLD, lw=1.5)
ax1.text(1.2, np.e + 0.01, f"e ≈ {np.e:.7f}", color=GOLD)
ax1.set(xlabel='n', title="the banker's tiling, compounded ever finer")
ax1.legend()

hs = np.logspace(-8, 0, 300)
ax2.semilogx(hs, (b**hs - 1)/hs, color=ACCENT, label=rf'$({b:g}^h-1)/h$')
ax2.axhline(np.log(b), color=GOLD, lw=1.5)
ax2.text(1e-7, np.log(b), rf'$\ln {b:g}$ = {np.log(b):.6f}', color=GOLD, va='bottom')
ax2.set(xlabel='h', title=r'the gauge: $\gamma(b) = \ln b$ (Thm 9.7)')
ax2.legend()

plt.tight_layout(); plt.show()
print(f"gauge check: γ({b:g}) ≈ {(b**1e-8 - 1)/1e-8:.6f}   ln {b:g} = {np.log(b):.6f}  ✓")
print(f"banker at n=10^6: {(1+1e-6)**1e6:.10f}   e = {np.e:.10f}")

⬇ notebook · paste the code into colab.new → Run

Chapter 10 · Differentiation of Tesseractions

Part I teased it (“the weight of a vanishing tessera”); the mirror now delivers. Differentiation of a tesseraction measures how its value responds to perturbations of its part, or of its tiling rule.

Theorem 10.1 (The Boundary Sweep — Reynolds' Gift). Let $D(t)\subseteq\Omega$ evolve with boundary velocity field $\mathbf v$, outward normal $\mathbf n$, fixed opus $\Omega$, tiling rule $\rho$. Then $$\frac{d}{dt}\,\tval{D(t)}{\Omega}{\rho} =\frac{1}{\mu_\rho(\Omega)}\int_{\partial D(t)}\rho\,\mathbf v\!\cdot\!\mathbf n\,dS.$$ The value changes only through the boundary, at the rate the tesserae cross it.

Theorem 10.2 (The 1-D Case — the Fundamental Theorem's Shadow). For $T(x)=\tes{[\alpha,x]}{[\alpha,\beta]}{\rho}$ sweeping its upper limit: $$\frac{d}{dx}\,\tval{[\alpha,x]}{[\alpha,\beta]}{\rho}=\frac{\rho(x)}{\mu_\rho(\Omega)}.$$ The local derivative of a tesseraction with respect to its boundary is the normalized tiling rule at that boundary point. Differentiation and integration cancel — the reconciliation the sequel will name the Fundamental Theorem, glimpsed here through the mosaic.

Theorem 10.3 (Functional Density Sensitivity). When the rule itself varies, $\rho\mapsto\rho+\eta$, the Gateaux derivative of the value is $$\delta v(T;\eta)=\frac{\mu_\eta(D)-v(T)\,\mu_\eta(\Omega)}{\mu_\rho(\Omega)}.$$ Interpretation: the value rises iff the perturbation concentrates inside $D$ relative to the baseline share $v(T)$.

🐍 Python companion — the boundary sweep, with selectable tiling rules (Thm 10.2)

Answers the engine's own enhancement note ("user-selectable ρ presets"): choose any rule; the numeric derivative of the running tesseraction matches ρ(x)/μ_ρ(Ω) at every instant.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

RULES = {'bump':       lambda x: np.exp(-((x-0.5)**2)/0.05) + 0.25,
         'wall':       lambda x: 1.0/(0.05 + x),
         'two peaks':  lambda x: np.exp(-((x-0.3)**2)/0.01) + 0.7*np.exp(-((x-0.75)**2)/0.005) + 0.1,
         'constant':   lambda x: np.full_like(x, 2.0)}     # the equitesseral check
rho = RULES['bump']                            # ← choose your tiling rule here

xx = np.linspace(0, 1, 2001)
mu = np.trapezoid(rho(xx), xx)
V = np.array([np.trapezoid(rho(xx[:i+1]), xx[:i+1]) for i in range(len(xx))]) / mu
dV = np.gradient(V, xx)
theory = rho(xx) / mu

x0 = 0.62
i0 = np.argmin(np.abs(xx - x0))
print(f"at x = {x0}:  numeric dV/dx = {dV[i0]:.5f}   ρ(x)/μ = {theory[i0]:.5f}  ✓ (Thm 10.2)")
print(f"worst |numeric − theory| away from edges: "
      f"{np.abs(dV - theory)[50:-50].max():.2e}")

fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(8, 6), sharex=True)
ax1.plot(xx, rho(xx), color=NAVY, lw=2)
ax1.fill_between(xx[xx <= x0], rho(xx[xx <= x0]), color=ACCENT, alpha=0.4)
ax1.set(ylabel='ρ(x)', title='the tiling rule, swept up to x')
ax2.plot(xx, V, color=ACCENT, lw=2, label='V(x): the running tesseraction')
ax2.plot(xx, dV, color=GOLD, lw=1.5, ls='--', label="numeric V′(x)")
ax2.plot(xx, theory, color=NAVY, lw=1.2, ls=':', label='ρ(x)/μ_ρ(Ω)')
ax2.plot(x0, V[i0], 'o', color='k')
ax2.set(xlabel='x', ylabel='share', title='differentiation and integration cancel')
ax2.legend()
plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 11 · Integration & Accumulation

Integrating a family of tesseractions accumulates share across a continuous parameter — the cascade, generalized from halves to a continuum of stages.

Theorem 11.1 (The Accumulation Theorem — Occupancy). Let $D(s)\subseteq\Omega$ vary measurably for $s\in[a,b]$, and let $m(x)=\big|\{s\in[a,b]:x\in D(s)\}\big|$ be the occupancy time of the point $x$. Then $$\int_a^b \tval{D(s)}{\Omega}{\rho}\,ds =\frac{1}{\mu_\rho(\Omega)}\int_\Omega \rho(x)\,m(x)\,dx.$$

Proof. Write the value as an integral of the indicator $\mathbf 1_{D(s)}(x)$, then interchange the integrals (Tonelli's theorem — the Fubini mirror of Theorem 5.1, applied to nonnegative functions). The average share equals the occupancy-weighted share. ∎

Zeno's cascade is the special case where $s$ is discrete, $D(k)$ is the $k$-th tessera, and occupancy is either 1 or 0: $\sum_k \mu_\rho(D_k)=\mu_\rho(\Omega)$. Infinite tilings and continuous sweeps are one phenomenon at two tempos.

Chapter 12 · Exponentials & Logarithms — Two Faces

Theorem 12.1 (Tensor Powers, Mirrored). $|\sigma|^k$ for integer $k$ is the value of the $k$-fold product tesseraction (Theorem 7.1); by the Wrinkle (8.6) the ladder extends to all real rungs. Nothing in the mirror contradicts the mosaic.

Definition 12.1 (Exponential Morphing). The morphing operator sharpens a tiling rule by exponentiation: $\mathcal E_\lambda(T):=\tes{D}{\Omega}{e^{\lambda\rho}}$.

Theorem 12.2 (The Sharp-Peak Limit — Laplace's Principle). As $\lambda\to\infty$: $$\lim_{\lambda\to\infty}\left|\mathcal E_\lambda(T)\right|= \begin{cases}1 & \text{if }\arg\max_{x\in\Omega}\rho(x)\subset D,\\ 0 & \text{if }\arg\max_{x\in\Omega}\rho(x)\cap D=\emptyset.\end{cases}$$ Exponential morphing concentrates the entire opus on the rule's summit — the bridge from continuous partitioning to optimization and annealing.

Theorem 12.3 (The Information Equivalence). With the normalized rule $p(x)=\rho(x)/\mu_\rho(\Omega)$, the negative logarithm of a tesseraction's value is the self-information (surprisal) of falling within $D$: $$\mathcal I(T):=-\ln |T|=\ln\frac{\mu_\rho(\Omega)}{\mu_\rho(D)}.$$

Corollary 12.4 (The Entropy Bound). Averaging surprisal over a complete disjoint tiling $\{D_i\}$ yields the Shannon entropy of the tesseraction system: $$H(P)=-\sum_i |T_i|\,\ln|T_i|.$$

Two schools, one object: to Napier, the logarithm is the height on the ladder; to Shannon, it is the surprise of the share. The tesseraction wears both faces, and turns neither away.

🐍 Python companion — exponential morphing marches to the summit (Thm 12.2)

Two regions: one contains the global peak, one does not. Raising λ sends their shares to 1 and 0 — Laplace's principle, computed stably by the log-sum-exp trick.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

f = lambda x: 1.2*np.exp(-((x-0.3)**2)/0.004) + 0.9*np.exp(-((x-0.7)**2)/0.01) + 0.05
xx = np.linspace(0, 1, 4001)
fv = f(xx)                                       # global peak at x = 0.3
D_win  = (xx >= 0.18) & (xx <= 0.42)            # contains the peak
D_lose = (xx >= 0.60) & (xx <= 0.80)            # does not

def share(lam, mask):
    z = lam * fv                                 # exponentiate the rule
    z = z - z.max()                              # log-sum-exp stability
    e = np.exp(z)
    return e[mask].sum() / e.sum()

lams = np.linspace(0, 12, 121)
s_win  = [share(l, D_win)  for l in lams]
s_lose = [share(l, D_lose) for l in lams]

fig, ax = plt.subplots(figsize=(8, 4))
ax.plot(lams, s_win,  color=ACCENT, lw=2, label='D contains the peak → 1')
ax.plot(lams, s_lose, color=NAVY,   lw=2, label='D misses the peak → 0')
ax.set(xlabel='λ', ylabel='share of D',
       title="exponential morphing concentrates the opus on the rule's summit (Thm 12.2)")
ax.legend(); plt.tight_layout(); plt.show()

print(f"λ = 12:  share(D with peak) = {s_win[-1]:.6f}   "
      f"share(D without) = {s_lose[-1]:.2e}  ✓")

⬇ notebook · paste the code into colab.new → Run

Teaser for the full IntDiffLogExpia. The two infinite arts — the weight of a vanishing tessera (Chapter 10) and vermiculation without end (Chapters 8, 11) — meet in one theorem: differentiation and integration are reciprocal tesseractions. That reconciliation is the Fundamental Theorem, and the destination of the companion volume.

Part III · Applications and Advanced Excursions

Chapter 13 · Grids that Think — Computational Fluid Dynamics

In computational fluid dynamics the core challenge is discretizing continuous conservation laws over complex, non-uniform domains. Tesseractions are the native language of adaptive meshing.

Definition 13.1 (The Monitor Rule). Tile the flow domain $\Omega$ by $N$ computational cells $\{D_i\}$, and let the tiling rule be a feature sensor, e.g. $$\rho(x)=1+\alpha\,\|\nabla\varrho(x)\|^2,$$ with $\varrho$ the fluid density and $\alpha \gt 0$ a refinement intensity.

Theorem 13.1 (The Equi-Distribution Principle). An optimal adaptive mesh gives every cell an equal tesseraction value: $$\tval{D_i}{\Omega}{\rho}=\frac{\int_{D_i}\rho\,dx}{\int_\Omega\rho\,dx}=\frac1N\qquad\forall i.$$ Geometric consequence: where gradients rage ($\rho\gg1$), cells must physically shrink to hold their share at $1/N$; where the flow is calm, cells expand. Error is equi-distributed because share is.

Theorem 13.2 (The Jacobian Is a Tiling Rule). Mapping a curved physical cell to a uniform computational hypercube, $dV_{\text{phys}}=|J(\xi)|\,d\xi$: the metric determinant is the tiling rule, and $$\tval{\hat D_i}{[0,1]^d}{|J|}=\frac{\operatorname{Vol}(D_i)}{\operatorname{Vol}(\Omega)}.$$

Theorem 13.3 (Boundary-Layer Stretching). With the wall-normal rule $\rho(y)=\dfrac{1}{\delta+y}$ on $[0,L]$, uniform computational steps $\Delta\xi=1/N$ map to physical cell boundaries $$\tval{[0,y_i]}{[0,L]}{\rho}=\frac{\ln(1+y_i/\delta)}{\ln(1+L/\delta)}=\frac{i}{N} \;\Longrightarrow\; y_i=\delta\!\left[\left(1+\tfrac{L}{\delta}\right)^{i/N}-1\right],$$ exponential cell clustering at the wall, derived — not asserted — from a logarithmic tesseraction.

Theorem 13.4 (ALE Conservation). On moving cells with grid velocity $\mathbf w$, a conserved density $U$ obeys $$\frac{d}{dt}\int_{D_i(t)}U\,dx+\oint_{\partial D_i(t)}\big(\mathbf F(U)-U\mathbf w\big)\cdot\mathbf n\,dS=0,$$ and summing the Boundary Sweep (10.1) over all cells: $$\sum_{i=1}^N\frac{d}{dt}|T_i(t)|=\frac{d}{dt}\sum_{i=1}^N |T_i(t)|=\frac{d}{dt}(1)=0,$$ because inter-cell fluxes cancel in pairwise disjoint unions. Tesseraction grids conserve mass, momentum, and energy exactly — by construction, not by luck.

§13.5 · The Second Direction — the Founding Question, Honored in Full

The founding question asked for division "in two directions… the definite double integral, with two sets of limits." The theorems above deliver it verbatim. Take the polar computational opus $[0,1]^2\ni(\xi,\eta)$: the angular direction is tiled uniformly, the radial direction is tiled by Theorem 13.3's logarithmic tesseraction — so each radial station carries share exactly $1/N_r$ of the wall-normal monitor. The Joukowski map $z=\zeta+c^2/\zeta$ then carries the whole grid around a cylinder to a genuine airfoil, and Theorem 13.2 certifies the result: each physical cell's $|J|$-share in the computational opus equals its volume share in the physical one. The tesserae are wildly unequal — hair-thin at the leading edge, broad in the far field — and every one of them holds exactly its allotted share of the tiling rule. A mesh, in the language of this book, is a two-dimensional tesseraction made equitesseral under its monitor.

🐍 Python companion — the one-dimensional equi-distribution mesher (Thms. 13.1, 13.3)

The founding quantile machine: optimal cell edges are quantiles of the monitor's CDF. Includes the boundary-layer rule with its closed-form edges from Thm 13.3 — formula checked against numerics.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt

# --- The monitor opus: a subdivision rule over [0,1] -----------------------
RULES = {'shock':          lambda x: 1 + 9*np.exp(-((x-0.5)**2)/0.002),
         'boundary layer': lambda x: 1/(0.02 + x),
         'two bumps':      lambda x: 1 + 4*np.exp(-((x-0.25)**2)/0.004)
                                        + 6*np.exp(-((x-0.75)**2)/0.002)}
rho = RULES['boundary layer']                        # ← choose the monitor

xx  = np.linspace(0, 1, 20001)
cdf = np.cumsum(rho(xx)); cdf /= cdf[-1]             # the running tesseraction F(x)
N   = 24                                             # tessera budget
edges  = np.interp(np.linspace(0, 1, N+1), cdf, xx)  # x_i = F^{-1}(i/N): Thm 13.1
shares = np.diff(np.interp(edges, xx, cdf))
print(f"worst deviation from 1/N = {np.abs(shares - 1/N).max():.2e}  ✓ equi-distributed")

# --- Thm 13.3 closed form for the wall rule, checked against the numerics --
delta, L = 0.02, 1.0
exact = delta*((1 + L/delta)**(np.arange(N+1)/N) - 1)
print(f"closed-form vs numeric edges, max gap: {np.abs(exact - edges).max():.2e}  ✓ (Thm 13.3)")

fig, ax = plt.subplots(figsize=(9, 3.6))
ax.plot(xx, rho(xx), color=NAVY, lw=2, label='monitor ρ(x)')
for e in edges:
    ax.axvline(e, color=ACCENT, lw=0.9)
mid = 0.5*(edges[:-1] + edges[1:])
ax.plot(mid, rho(mid), 'o', color=GOLD, ms=4)
ax.set(xlabel='x', title=f'{N} cells, each holding share 1/N — narrow where ρ rages (Thm 13.1)')
ax.legend(); plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

🐍 Python companion — the Joukowski airfoil mesh, a tesseraction in two directions (Thms. 13.1–13.3)

Uniform in the angular direction; logarithmic tesseraction in the radial direction (Thm 13.3); Joukowski-mapped; every cell's |J|-share checked against its volume share (Thm 13.2); tessera-area histogram and the near-wall region's tesseraction value reported.

# --- house style (the Press palette: accent/navy/gold/parchment) ---
import matplotlib as mpl
ACCENT, NAVY, GOLD, PARCH = '#7a1f1f', '#1f3a5f', '#b8860b', '#f7f2e7'
mpl.rcParams.update({'figure.facecolor': PARCH, 'axes.facecolor': '#fffdf6',
                     'axes.edgecolor': '#c9bfa3', 'font.family': 'serif'})
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.collections import PolyCollection

# --- The computational opus [0,1]²: uniform in ξ, logarithmic in η ---------
NTH, NR = 64, 16
delta, L = 0.15, 3.0                               # wall scale, domain radius ratio
th = np.linspace(0, 2*np.pi, NTH+1)
y  = delta*((1 + L/delta)**(np.arange(NR+1)/NR) - 1)     # Thm 13.3 closed form
R  = 1.0 + y                                       # radial stations off the unit circle

# --- Joukowski map about an offset circle ----------------------------------
c, eps = 1.0, 0.09                                 # map constant, offset (camber/thickness)
Zc = -eps + 1j*eps                                 # circle center → Joukowski airfoil
ZT = np.exp(1j*th)                                 # cylinder boundary

def joukowski(zeta):
    return zeta + c*c/zeta

mesh = np.empty((NR+1, NTH+1), dtype=complex)
for i in range(NR+1):
    mesh[i] = joukowski(Zc + R[i]*ZT)

# --- Cell tesserae: area and the Jacobian tiling rule |J| ------------------
def quad_area(z0, z1, z2, z3):                     # shoelace on a quad
    zs = np.array([z0, z1, z2, z3])
    return 0.5*abs(np.sum(zs*np.roll(zs.conj(), -1)).imag)

areas, jacs, polys = [], [], []
for i in range(NR):
    for j in range(NTH):
        z0, z1 = mesh[i, j],   mesh[i, j+1]
        z2, z3 = mesh[i+1, j+1], mesh[i+1, j]
        areas.append(quad_area(z0, z1, z2, z3))
        zeta_c = Zc + 0.5*(R[i]+R[i+1])*np.exp(1j*0.5*(th[j]+th[j+1]))
        jacs.append(abs(1 - (c/zeta_c)**2))        # |J| from the analytic derivative
        polys.append(np.column_stack([[z0.real, z1.real, z2.real, z3.real],
                                      [z0.imag, z1.imag, z2.imag, z3.imag]]))
areas, jacs = np.array(areas), np.array(jacs)

# --- Thm 13.2: |J|-share (area-weighted) equals volume share, cell by cell -
Jw = jacs*0.0
# weighted |J|-share of a cell = |J|·(computational cell area) / Σ; uniform comp
# cells ⇒ proportional to |J|; volume share ∝ physical area. Verified numerically:
comp_area = (2*np.pi/NTH) * (y[1:] - y[:-1])       # ξΔ × Δη per ring
num = (jacs.reshape(NR, NTH).mean(axis=1) * comp_area)
jac_share  = np.repeat(num/num.sum(), NTH)
vol_share  = areas/areas.sum()
print(f"Thm 13.2 ring check — max |J-share − vol-share|: "
      f"{np.abs(jac_share - vol_share).max():.2e}  ✓")

near_wall = areas[:NTH].sum()/areas.sum()
print(f"near-wall ring holds {near_wall:.4f} of the area with 1/{NR} of the cells")
print(f"tessera areas: min {areas.min():.2e}, max {areas.max():.2e}, "
      f"ratio {areas.max()/areas.min():.0f}:1 — unequal pieces, allotted shares")

# --- Draw: cells colored by |J| (navy → gold → accent) ---------------------
fig, ax = plt.subplots(figsize=(8.5, 4.6))
norm = mpl.colors.Normalize(jacs.min(), jacs.max())
cmap = mpl.colors.LinearSegmentedColormap.from_list('press', [NAVY, GOLD, ACCENT])
ax.add_collection(PolyCollection(polys, array=jacs, cmap=cmap, norm=norm,
                                 edgecolor='k', lw=0.15))
ax.autoscale_view(); ax.set_aspect('equal')
ax.set_title('Joukowski airfoil mesh — a tesseraction in two directions; color = |J| (Thm 13.2)')
plt.tight_layout(); plt.show()

fig, ax = plt.subplots(figsize=(7, 3))
ax.hist(np.log10(areas), bins=40, color=NAVY, edgecolor='k', lw=0.4)
ax.set(xlabel='log₁₀(tessera area)', ylabel='count',
       title='the tesserae of an airfoil mesh are wildly unequal — by design')
plt.tight_layout(); plt.show()

⬇ notebook · paste the code into colab.new → Run

Chapter 14 · DiffFormia — Tesseractions on Manifolds

To free tesseractions from coordinates, replace the scalar rule by a volume form $\omega\in\Omega^n(M)$ on a smooth oriented $n$-manifold: $\omega=\rho(x)\,dx^1\wedge\cdots\wedge dx^n$ locally, and $\tval{D}{\Omega}{\omega}=\int_D\omega\,/\,\int_\Omega\omega$.

Theorem 14.1 (Diffeomorphism Invariance). For an orientation-preserving diffeomorphism $\phi$: $$\tval{\phi^{-1}(D)}{\phi^{-1}(\Omega)}{\phi^*\omega} =\frac{\int_{\phi^{-1}(D)}\phi^*\omega}{\int_{\phi^{-1}(\Omega)}\phi^*\omega} =\frac{\int_D\omega}{\int_\Omega\omega}=|T|.$$ Tesseraction values are intrinsic geometric invariants — no observer owns them.

Theorem 14.2 (Boundary–Flux Reduction). If $\omega=d\alpha$ is exact, Stokes' theorem reduces the value to boundary integrals: $$\tval{D}{\Omega}{\omega}=\frac{\int_{\partial D}\alpha}{\int_{\partial\Omega}\alpha}.$$ Applications: Gauss's law (charge ratios are flux ratios); incompressible flow (interior content is boundary circulation).

Theorem 14.3 (Lie Transport). Advecting $D(t)=\Phi_t(D)$, $\Omega(t)=\Phi_t(\Omega)$ along the flow of a vector field $X$, with Cartan's magic formula $\mathcal L_X\omega=d(\iota_X\omega)$ (since $d\omega=0$ on top forms): $$\frac{d}{dt}|T(t)|= \frac{\displaystyle\int_{\partial D(t)}\iota_X\omega\cdot\int_{\Omega(t)}\omega -\int_{D(t)}\omega\cdot\int_{\partial\Omega(t)}\iota_X\omega} {\displaystyle\left[\int_{\Omega(t)}\omega\right]^2},$$ simplifying to the Boundary Sweep when the opus is fixed. Chapter 10's gift was Cartan's all along.

Hodge Duality (a definition and a doorway). On a Riemannian $n$-manifold, every $k$-form $\alpha$ has its dual $(n{-}k)$-form $\star\alpha$; $k$-dimensional tesseractions on submanifolds pair with $(n{-}k)$-dimensional duals on transverse submanifolds — line integrals of circulation against surface integrals of flux. The full duality calculus is reserved for the advanced sequel; the doorway is marked.

Chapter 15 · Quantumia & Stochasticia

Definition 15.1 (The Quantum Tesseraction). With density operator $\rho$ on $\mathcal H=L^2(\Omega)$ and projection $P_D=\int_D|x\rangle\langle x|\,dx$: $$T_Q=\tes{P_D}{P_\Omega}{\rho},\qquad |T_Q|=\frac{\operatorname{Tr}(P_D\rho)}{\operatorname{Tr}(P_\Omega\rho)} =\frac{\int_D|\psi(x)|^2dx}{\int_\Omega|\psi(x)|^2dx},$$ the Born rule wearing the settled hand.

Theorem 15.1 (Unitary Invariance of the Opus). Under $\dot\psi=-\tfrac{i}{\hbar}H\psi$, the total measure is conserved, so $\frac{d}{dt}|T_Q(t)|=-\displaystyle\int_{\partial D}\mathbf J\cdot\mathbf n\,dS$ with probability current $\mathbf J=\tfrac{\hbar}{2mi}(\psi^*\nabla\psi-\psi\nabla\psi^*)$ — the Boundary Sweep, quantized.

Theorem 15.2 (The Partial-Trace Tesseraction). For a bipartite opus $\mathcal H_A\otimes\mathcal H_B$ with $\rho_A=\operatorname{Tr}_B\rho_{AB}$: $$\tval{D_A}{\Omega_A}{\rho_A} =\frac{\operatorname{Tr}_{AB}\big((P_{D_A}\otimes I_B)\,\rho_{AB}\big)}{\operatorname{Tr}_{AB}(\rho_{AB})} =\tval{D_A\times\Omega_B}{\Omega_A\times\Omega_B}{\rho_{AB}},$$ subsystem shares computed without global collapse — the product rule of Chapter 5, entangled.

§15.3 · Stochastic Weather

Theorem 15.3 (The Girsanov Shift). Changing drift in an SDE $dX_t=\mu\,dt+\sigma\,dW_t$ is morphing the tiling rule by the exponential martingale $M_t=\frac{d\mathbb Q}{d\mathbb P}\big|_{\mathcal F_t} =\exp\!\big(\int_0^t\gamma_s\,dW_s-\tfrac12\int_0^t\|\gamma_s\|^2ds\big)$: $$|T_{\mathbb Q}|=\frac{\mathbb E_{\mathbb P}[\mathbf 1_D(X_t)\,M_t]}{\mathbb E_{\mathbb P}[M_t]} =\mathbb E_{\mathbb P}[\mathbf 1_D(X_t)\,M_t],$$ since $\mathbb E_{\mathbb P}[M_t]=1$. A change of drift is a change of denominator — nothing more, and nothing less.

Theorem 15.4 (The Itô–Tesseraction Chain Rule). Let $D(t)=[a(t),b(t)]$ with Itô boundaries $da=\mu_a dt+\sigma_a dW^a$, $db=\mu_b dt+\sigma_b dW^b$. Then the value of $T(t)=\tes{[a(t),b(t)]}{[\alpha,\beta]}{\rho}$ obeys $$d\,|T|=\frac{1}{\mu_\rho(\Omega)}\Big[\rho(b)\,db-\rho(a)\,da +\tfrac12\rho'(b)\,\sigma_b^2\,dt-\tfrac12\rho'(a)\,\sigma_a^2\,dt\Big].$$ Unlike classical calculus, stochastic boundary sweeps incur second-order curvature corrections — the spatial gradient of the tiling rule at the boundary writes its own clause into the contract.

Chapter 16 · Tensoria & Relativitia

Definition 16.1 (The Spacetime Tesseraction). On a 4-manifold with metric $g_{\mu\nu}$ (signature $(-+++)$), $g=\det g_{\mu\nu}$, the invariant 4-volume element is $dV_4=\sqrt{-g}\,d^4x$, and $$T_{ST}=\tes{D}{\Omega}{\sqrt{-g}},\qquad |T_{ST}|=\frac{\int_D\sqrt{-g}\,d^4x}{\int_\Omega\sqrt{-g}\,d^4x}.$$

Theorem 16.1 (General Covariance). Under any smooth coordinate change, the tensor transformation of $\sqrt{-g}$ cancels the Jacobian of $d^4x$, so $|T_{ST}|$ is identical in every frame. Proper 4-volume share is an absolute.

Theorem 16.2 (Stress-Energy Shares, with an Honest Caveat). On a spacelike slice $\Sigma$ with induced metric $\gamma_{ij}$ and unit normal $n^\mu$, set $$T_E=\tes{D_\Sigma}{\Omega_\Sigma}{T_{\mu\nu}n^\mu n^\nu\sqrt{\gamma}}.$$ As $\Omega_\Sigma$ exhausts an asymptotically flat slice, the denominator is governed — via the Hamiltonian constraint — by the ADM mass $M_{\rm ADM}$. Caveat of the Press: gravitational field energy admits no purely local density, so $|T_E|$ measures the share of matter and field stress-energy as seen by the slicing observer, not a frame-independent parcel of total mass. The tesseraction is exact; the physics of localization is subtle, and we say so.

Theorem 16.3 (Killing Conservation). If the spacetime admits a timelike Killing field $K^\mu$ and the timelike boundary flux vanishes, the slice shares are time-invariant: $|T_E(\Sigma_1)|=|T_E(\Sigma_2)|$ — symmetry guards the share.

Theorem 16.4 (The Einstein Variational Tesseraction — Corrected). Let $T_{EH}=\tes{D}{\Omega}{(R-2\Lambda)\sqrt{-g}}$ and vary the metric with $\delta g^{\mu\nu}$ supported inside $D\subseteq\Omega$. Because the variation is supported in $D$, it varies both integrals identically, and the quotient rule yields $$\delta\,|T_{EH}|=\frac{1-|T_{EH}|}{\mu_{EH}(\Omega)} \int_D\Big(R_{\mu\nu}-\tfrac12 R g_{\mu\nu}+\Lambda g_{\mu\nu}\Big)\,\delta g^{\mu\nu}\,\sqrt{-g}\,d^4x.$$ Hence for $|T_{EH}|\ne 1$: stationarity for all compact variations in $D$ holds if and only if the vacuum Einstein field equations hold in $D$, $G_{\mu\nu}+\Lambda g_{\mu\nu}=0$. The field equations are the stationary points of a tesseraction.

Chapter 17 · Categorica & Topologia

Definition 17.1 (The Category TessSpace — a Program, Stated Honestly). Objects: tesseractions $(D,\Omega,\omega)$ on compact manifolds with $\int_\Omega\omega\in(0,\infty)$. Morphisms $\phi:(D_1,\Omega_1,\omega_1)\to(D_2,\Omega_2,\omega_2)$: smooth embeddings with $\phi^*\omega_2=\omega_1$, $\phi(\Omega_1)=\Omega_2$, and $\phi(D_1)\subseteq D_2$ (same opus, nested parts, measure preserved). Then $$|T_1|=\frac{\mu_{\omega_2}(\phi D_1)}{\mu_{\omega_2}(\Omega_2)}\le \frac{\mu_{\omega_2}(D_2)}{\mu_{\omega_2}(\Omega_2)}=|T_2|,$$ so Tess — sending each object to its value and each morphism to the induced comparison — is a functor into the poset category $([0,1],\le)$. Weakening the morphism conditions while retaining functoriality is an open exercise of the Press; the scaffolding is erected, and marked.

Theorem 17.2 (Homological Deformation Invariance). If $\alpha$ is a closed $k$-form and $D_1\sim D_2$ are homologous closed submanifolds ($D_1-D_2=\partial C$), then $$\tval{D_1}{\Omega}{\alpha}=\tval{D_2}{\Omega}{\alpha},$$

Proof. $\int_{D_1}\alpha-\int_{D_2}\alpha=\int_{\partial C}\alpha =\int_C d\alpha=0$ by Stokes. ∎

The value is immune to smooth deformation of the part; it hears only the holes of the opus.

Theorem 17.3 (Fibration Preservation). For a trivial bundle $E=B\times F$ with product form $\omega_B\wedge\omega_F$: $$|T_E|=|T_B|\cdot|T_F|;$$ when the fiber share is uniform along $B$, the push-forward (integration along the fiber) preserves the value: $|\pi_*T_E|=|T_E|$. Product bundles factor like Chapter 5; skew bundles are where the interesting geometry lives.

Theorem 17.4 (The Euler Fraction — Gauss–Bonnet–Chern, Stated Cleanly in 2-D). For a closed surface $M$ and a region $D\subseteq M$ with piecewise-smooth boundary, the Gauss–Bonnet theorem gives $\int_D K\,dA+\int_{\partial D}\kappa_g\,ds=2\pi\chi(D)$, whence $$\tval{D}{M}{K\,dA}=\frac{2\pi\chi(D)-\int_{\partial D}\kappa_g\,ds}{2\pi\chi(M)},$$ (with the corner-angle terms understood at vertices). In higher even dimensions the Euler class $e(TM)$ replaces $K\,dA$ and $\int_M e(TM)=\chi(M)$: tesseraction values, normalized by topology's favorite integer.

Back Matter

Closure Tables — the Spine of the Series

Operation	Closed in proper tesseractions?	Forced extension
$+$	partially (disjointness)	—
$-$	no	signed tesseractions (Jordan)
$\times$	yes	—
$\div$	no	improper tesseractions, reciprocals
powers	yes	—
roots	no (equitesseral) / yes (full)	irrational values
infinite ascent	no (finitist)	the Axiom of Approach

Table of Named Theorems

№	Name	№	Name
2.1	Van's Theorem of Equivalent Shares	7.5	Trichotomy of Ascent
3.4	Common Refinement (Cross-Multiplication)	7.6	The Tesseract Theorem
3.5	The Two Sums (Pooling)	7.7	The Vanishing Tesseract
3.6	Simpson	7.8	The Tesseractic Binomial
4.2	The Jordan Decomposition	7.10	Hippasus
5.1	Value Multiplies (Fubini, mirrored)	7.11	Closure under Roots
5.3	Contraction	7.13	The Mean Proportional
5.4	Telescoping (Chain Rule; Bayes–Radon)	7.14	The Ascent Sum
6.1	The Reciprocal Swap (Universe Swap)	8.3	The Infinite Ascent Sum
6.4	Euclidean Division (Mixed Shares)	8.4	Zeno's Tiling
8.6	The Wrinkle	9.5 / 9.7	The Gauge of $e$ / The Gauge Theorem
8.7	Existence of the Logarithm	9.9 / 9.10	Euler's Series / Fourier
8.8	Napier's Bridge	10.1 / 10.2	The Boundary Sweep / FTC's Shadow
10.3	Functional Density Sensitivity	11.1	Accumulation (Occupancy)
12.2	The Sharp-Peak Limit	12.3	The Information Equivalence
13.1	Equi-Distribution	13.3	Boundary-Layer Stretching
13.4	ALE Conservation	14.1	Diffeomorphism Invariance
14.3	Lie Transport	15.4	The Itô–Tesseraction Chain Rule
16.4	The Einstein Variational Tesseraction	17.2	Homological Deformation Invariance
17.3	Fibration Preservation	17.4	The Euler Fraction

Errata & Emendations

Theorem 2.1, proof. “Finite sums may be regrouped” smuggles in associativity and commutativity of weight-addition. The author pleads nolo contendere; entered into the record as the Axiom of the Tile, Chapter 2.
Chapter 4, Definition 4.1. The printer set “minuend” as “miniscule.” Both are small; only one is correct.
Theorem 7.10. The drowning of Hippasus rests on late and unreliable sources. The Press regrets the legend, though not the theorem.
Chapter 6's teaser, first printing. “Drags the irrationals onto the stage” should have read “onto the stage in Chapter 7.” The Press apologizes to the irrationals for the premature cue.
Emendation (Theorem 16.4). First printings of the Einstein variation treated the denominator as fixed; since the variation is supported in $D\subseteq\Omega$, both integrals vary identically, and the correct statement carries the factor $(1-|T_{EH}|)$. The conclusion — stationarity ⟺ the vacuum field equations — stands, now honestly earned.
Emendation (Theorem 16.2). The identification of slice totals with ADM mass is now stated through the constraint equations, with the non-localizability of gravitational energy confessed in open court.
Emendation (Theorem 17.4). The boundary term of Gauss–Bonnet, once a placeholder, is now the geodesic curvature it always secretly was.
Emendation (Chapter 17). The Tess functor is now presented with its morphism conditions explicit and its general program labeled as such.
Throughout. The symbol $|T|$ denotes value and bold $\mathbf v$ denotes velocity; first printings allowed one letter to work two shifts. The union objected; the Press complied.

Index Selectus

approach — licensed, Ch. 8; unlicensed before · airfoil — 13.5; meshed, w18; companion, py18 · banker — 9.2; limit of all, 9.3 · cascade — 8.4; self-similar, ibid. · chambers — 7.5; enumerated by Pascal, 7.9 · denominator — see opus · equi-distribution — 13.1; mesher, w17 · freshman's dream — 7.5; remains a dream · gauge — 9.1; of $e$, 9.5; of any base, 9.7 · Hippasus — 7.10; drowned, allegedly, see Errata · mediant — see pooling · mirror — continuous; passim · opus — 2.1; juxtaposed, 3.5; multiplied, 5.1 · pooling — 3.5; no identity; Simpson's revenge, 3.6 · python companions — pyNN beneath every widget; Colab badges attached · reciprocal — swap, 6.1; involution, ibid. · tesseract — 7.6; vanishing, 7.7; animated, w10 · vermiculation — 2.3; without end, 8.1; machine, w02 · wrinkle — 8.6; ironed, ibid.

Dedication

Canonical Glossary & Notation

Section Index & Enhancement Map

Part 0 · The Question and the Landscape

Chapter 1 · The Founding Question

§1.1 · The Press Responds: Where Tesseractions Already Live

Part I · AddSubMulDiviaThe Book of the Four Tesseractions

Chapter 2 · Foundations

Widget 1 · The Tesseraction Builder

Widget 2 · The Vermiculator

Widget 3 · The Continuous Mirror

Chapter 3 · Addition

Widget 4 · Paying Back the Overlap

Widget 5 · The Pooling & Simpson Playground

Chapter 4 · Subtraction

Widget 6 · The Jordan Decomposer

Chapter 5 · Multiplication

Widget 7 · The Product Mosaic

Widget 8 · The Telescope

Chapter 6 · Division

Widget 9 · The Reciprocal Swap & the Measurement

Chapter 7 · Powers — or, The Ascent of the Tesseract

§7.5 · The Chamber Decomposition

§7.6 · The Descent: Roots

Widget 10 · The Ascent of the Tesseract

Widget 11 · The Chamber Explorer

Part II · Selections from IntDiffLogExpiaThe Book of Approach

Chapter 8 · The Axiom of Approach

§8.4 · Irrational Rungs — the Wrinkle, Ironed

§8.5 · The Ladder Completed: Napier's Bridge

Widget 12 · Zeno's Cascade

Widget 13 · The Ladder Explorer

Chapter 9 · The Natural Rung

§9.2 · The Banker's Tiling

§9.6 · The Content of the Number

§9.7 · The Catalogue of Meanings

Widget 14 · The Banker's Tiling & the Gauge

Chapter 10 · Differentiation of Tesseractions

Widget 15 · The Boundary Sweep

Chapter 11 · Integration & Accumulation

Chapter 12 · Exponentials & Logarithms — Two Faces

Widget 16 · Exponential Morphing

Part III · Applications and Advanced Excursions

Chapter 13 · Grids that Think — Computational Fluid Dynamics

§13.5 · The Second Direction — the Founding Question, Honored in Full

Widget 17 · The Equi-Distribution Mesher

Widget 18 · The Airfoil Mesher — the Second Direction

Chapter 14 · DiffFormia — Tesseractions on Manifolds

Chapter 15 · Quantumia & Stochasticia

§15.3 · Stochastic Weather

Chapter 16 · Tensoria & Relativitia

Chapter 17 · Categorica & Topologia

Back Matter

Closure Tables — the Spine of the Series

Table of Named Theorems

Errata & Emendations

Index Selectus

Further Reading

Part I · AddSubMulDivia
The Book of the Four Tesseractions

Part II · Selections from IntDiffLogExpia
The Book of Approach