Definite Integral as a Limit of Sums: Concepts and Examples

Before calculus had its shortcut formulas, mathematicians computed areas by adding up thousands of thin rectangles under a curve — the method of exhaustion used by Archimedes. The definite integral as a limit of sums is the modern version of that idea: slice the interval [a, b] into n thin strips, add their areas, and let n grow without bound so the strips become infinitely thin.

This definition matters because it is the true meaning of the definite integral — every property and application flows from it. It also gives a way to evaluate integrals of simple functions from first principles, and it is the foundation of numerical integration methods used by computers when no formula exists.

Key Concepts

• Partition: divide [a, b] into n equal subintervals, each of width h = (b − a)/n.
• Riemann sum: add f evaluated at strip positions: h·[f(a) + f(a + h) + f(a + 2h) + … + f(a + (n−1)h)].
• The limit: ∫ₐᵇ f(x) dx = lim (h → 0, n → ∞) of that sum.
• Sum formulas needed: 1 + 2 + … + (n−1) = (n−1)n/2; squares sum: 1² + 2² + … + (n−1)² = (n−1)n(2n−1)/6; cubes: [(n−1)n/2]².
• Geometric meaning: the sum approximates area with rectangles; the limit makes the approximation exact.

Formulas

• ∫ₐᵇ f(x) dx = lim(n→∞) h [f(a) + f(a+h) + … + f(a + (n−1)h)], h = (b−a)/n
• lim(n→∞) 1/n = 0; lim(n→∞) (n−1)/n = 1; lim(n→∞) (n−1)(2n−1)/n² = 2
• ∫₀ᵇ x dx = b²/2; ∫₀ᵇ x² dx = b³/3 (both provable from first principles)

Worked Example

Evaluate ∫₀² x² dx as a limit of sums. Step 1: h = (2 − 0)/n = 2/n. Step 2: the sum is h·[h² + (2h)² + … + ((n−1)h)²] = h³·[1² + 2² + … + (n−1)²] = (8/n³)·(n−1)n(2n−1)/6. Step 3: simplify: (8/6)·[(n−1)/n]·[(2n−1)/n] = (4/3)·(1 − 1/n)·(2 − 1/n). Step 4: let n → ∞: the bracketed factors tend to 1 and 2, giving (4/3)·1·2 = 8/3. This matches the Fundamental Theorem result (2³/3 − 0), confirming the definition.

Common Misconceptions

• Using n strips but forgetting h → 0: the limit requires both n → ∞ and h → 0 together; a finite sum is only an approximation.
• Off-by-one in the sum: the standard right-endpoint form stops at (n−1)h, not nh; miscounting terms breaks the limit.
• Thinking the definition is obsolete: the antiderivative shortcut only works when you can find one — numerical methods still use sums for integrals with no closed form.

Real-World Applications

Computers evaluate integrals they cannot do symbolically (like ∫ e^(−x²) dx) using refined Riemann sums — the trapezoidal and Simpson rules are its direct descendants. In finance, option-pricing integrals are computed numerically this way. Physicists use the sum picture to define work, mass and charge as limits of discrete pieces.

The definite integral is, at heart, a limit of sums — infinitely many, infinitely thin rectangles. The antiderivative shortcut is a wonderful convenience, but this definition is the concept itself, and it never fails.