JOVANA
Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
All guides

The Two Fundamental Theorems of Calculus

Prove both halves: differentiating the accumulated-area function gives f back, and an antiderivative lets you evaluate any integral by subtraction. This is where integration and differentiation become inverse operations.

The accumulation function

Let f be integrable on [a,b]. Define the accumulation function F(x) = ∫ from a to x of f(t) dt. This F measures the running area swept out as the right edge moves from a to x. The fundamental theorem of calculus, part one, says: if f is continuous at a point x, then F is differentiable there and F′(x) = f(x). Sweeping out area and then differentiating returns the original f.

FTC, Part 1.  f integrable on [a,b], continuous at x. Then F'(x) = f(x),
              where F(x) = integral from a to x of f.

Proof. Look at the difference quotient. For h > 0 (h < 0 is symmetric),
   F(x+h) - F(x) = integral from a to x+h  -  integral from a to x
                 = integral from x to x+h of f(t) dt      (additivity).
Subtract the constant f(x), written as integral from x to x+h of f(x) dt = f(x)*h:
   F(x+h) - F(x) - f(x)*h  =  integral from x to x+h of ( f(t) - f(x) ) dt.

Let ε > 0. Continuity at x: there is δ>0 with |t - x| < δ => |f(t) - f(x)| < ε.
For 0 < h < δ every t in [x, x+h] satisfies |t - x| < δ, so by |∫g| <= ∫|g|:
   | F(x+h) - F(x) - f(x)*h |  <=  integral from x to x+h of |f(t)-f(x)| dt
                              <=  ε * h.
Divide by h:   | (F(x+h) - F(x))/h  -  f(x) |  <=  ε.
Since ε was arbitrary, the difference quotient -> f(x). Hence F'(x) = f(x).  QED
FTC Part 1: the derivative of accumulated area is the integrand.

Evaluation by antiderivative

Part two is the one you actually compute with. If f is integrable on [a,b] and G is any antiderivative of f — meaning G′ = f everywhere on [a,b] — then ∫ from a to b of f = G(b) − G(a). You never touch a single partition; you find an antiderivative and subtract its endpoint values.

FTC, Part 2.  f integrable on [a,b], G'(x) = f(x) for all x in [a,b].
              Then  integral from a to b of f  =  G(b) - G(a).

Proof. Take ANY partition P: a = x_0 < ... < x_n = b. Telescope G across it:
   G(b) - G(a) = sum_{k=1}^n ( G(x_k) - G(x_{k-1}) ).
Apply the Mean Value Theorem to G on each [x_{k-1}, x_k]: there is a point
   t_k in (x_{k-1}, x_k) with  G(x_k) - G(x_{k-1}) = G'(t_k) Δx_k = f(t_k) Δx_k.
So   G(b) - G(a) = sum f(t_k) Δx_k  =  a Riemann sum of f for P.
But on each piece  m_k <= f(t_k) <= M_k, hence
   L(f,P) <= G(b) - G(a) <= U(f,P)   for EVERY partition P.
The only number squeezed between all lower and all upper sums is the integral:
   integral from a to b of f  =  G(b) - G(a).   QED
FTC Part 2 via the Mean Value Theorem and the squeeze between L and U.

A useful companion: the integral mean value theorem

A close relative is the mean value theorem for integrals: if f is continuous on [a,b], there is a point c in [a,b] with ∫ from a to b of f = f(c)·(b − a). Geometrically, the area equals that of a rectangle whose height is the function's value at some interior point — the average value f(c) is genuinely attained.

Integral MVT.  f continuous on [a,b]. Let m = min f, M = max f on [a,b].
Monotonicity of the integral gives
   m*(b-a) <= integral from a to b of f <= M*(b-a),
so the AVERAGE value  A = (1/(b-a)) * integral from a to b of f  lies in [m, M].
f is continuous and attains m and M (Extreme Value Thm), so by the
Intermediate Value Theorem it attains every value in between, including A.
Hence there is c in [a,b] with f(c) = A, i.e.
   integral from a to b of f = f(c)*(b - a).   QED
The integral mean value theorem from monotonicity plus IVT.