Accumulation and integrals

People needed areas of curves and totals built from rates long before school named the integral: distance from a changing speed, grain from a varying density, the region under a bow-shaped boundary that refuses to be a tidy rectangle. Derivatives measure change. Accumulation rebuilds the total.

Long before formal calculus, Archimedes exhausted areas and volumes with relentless approximation: trap the unknown between shapes he could control. MacTutor's biography is the standard entry. Steal the ethic: patience with limits. Do not demand the whole answer on the first polygon. Squeeze until only one possibility remains. Centuries later, integral calculus industrializes that boldness with systematic sums and a handshake with differentiation.

For a modern rereader, this chapter teaches you to rebuild totals from rates, then meets the fundamental theorem that makes calculus one subject rather than two clever tricks. Without that link, you have techniques. With it, you have a closed loop of ideas.

Slice an interval into thin widths $\Delta x$, take height about $f(x_i)$, and form area about $f(x_i)\Delta x$. Sum the slices. As slices thin, the sum approaches the definite integral $\int_a^b f(x)\,dx$ when the limit exists.

If $f$ is a rate, the integral accumulates net change. Signed area matters: when $f$ dips below zero, contributions subtract. Direction and deficit are not decoration.

A function $F$ with $F' = f$ is an antiderivative of $f$. Adding a constant does not change the derivative, so antiderivatives come in families $F + C$.

Fundamental theorem (intuition): differentiating an accumulation from a fixed start recovers the rate; integrating a rate recovers net accumulation. Derivative and integral are inverse stories under suitable conditions.

Each thin rectangle $f(x_i)\Delta x$ is a tiny account of accumulation. Add them. Refine the partition. The integral is what the sum is trying to become.

If you have ever approximated distance by reading a speedometer every few seconds and multiplying by time gaps, you have already lived a Riemann sum. Gaps matter. Sign matters if direction reverses. Sparse readings give a rough integral; dense readings give a better one. Calculus names the ideal limit of that refinement.

Density over volume becomes mass. Power over time becomes energy. Loss over a probability distribution becomes expected risk. The pattern is always: local contribution times tiny measure, then sum, then refine.

Algebrica's definite-integrals pages are diagram-forward companions. The fundamental theorem is the handshake between this lesson and the derivative pages.

Area always means "positive paint." Signed integrals record cancellation. Integrating velocity can return displacement (net), not total distance traveled. Total distance needs absolute speed, or careful splitting where the sign changes.

Antiderivatives without $+C$. Because constants vanish under differentiation, indefinite integrals name families. Forgetting $+C$ is forgetting that many heights share one slope function.

"Integral equals antiderivative" without limits. The definite integral $\int_a^b f$ is a number (net accumulation). An antiderivative is a function. The fundamental theorem connects them: evaluate an antiderivative at the endpoints (under the usual hypotheses).

Confusing average height with total accumulation. Average value of $f$ on $[a,b]$ is $\frac{1}{b-a}\int_a^b f$. The integral itself still has units of $f$ times $x$, not the units of $f$ alone.

Thinking every integral is geometric area. Area is the best picture, not the only meaning. Accumulation of rate, mass from density, expectation from weighted loss: same pattern, different units.

Ignoring units in $\Delta x$. The thin width carries units. Forget them and the sum's units become mystery meat.

Long before formal calculus, Archimedes exhausted areas and volumes with relentless approximation: trap the unknown between shapes he could control. Squeeze until only one possibility remains. MacTutor's Archimedes biography is the standard entry point. The ethic is patience with limits. Do not demand the whole answer on the first polygon.

Centuries later, integral calculus industrializes that boldness with systematic sums and the fundamental theorem. Next time you approximate an irregular lake on a map with grid squares, you are in the same family of mind.

Distance from speed over time, total energy from power over time, and total rainfall from rate charts are accumulation problems. Accounting systems that add contributions over bins are discrete cousins of the same idea.

A trip reconstructed from sparse GPS speeds is a Riemann sum in traffic. A continuous model is the dream of infinite refinement. Engineers live between those poles constantly.