Unit 6 · Topic 6.4 Beta

The Fundamental Theorem and Accumulation Functions

4 min read · freeNot practiced

So far a definite integral has been a single number: the area under ff from aa to bb. Now let the right end move. The area then depends on where you stop, so it is a function. Studying that function leads to the most important theorem in the course.

Accumulation functions

Fix a starting point aa and define

g(x)=∫axf(t)dt.g(x) = \int_a^x f(t)\,dt.

For each xx, g(x)g(x) is the amount accumulated under ff from aa up to xx. This is an accumulation function. The variable inside is renamed tt so it does not clash with the xx in the limit. Two facts follow at once:

  • g(a)g(a)=∫aaf(t)dt= \int_a^a f(t)\,dt=0= 0: no width, no area.
  • Values of gg come from areas. If ff is a graph of lines and arcs, use geometry.

Worked example 1. ff is made of segments from (0,1)(0, 1) to (2,3)(2, 3), from (2,3)(2, 3) to (4,3)(4, 3) and from (4,3)(4, 3) to (6,0)(6, 0). Let g(x)g(x)=∫0xf(t)dt= \int_0^x f(t)\,dt. Find g(2)g(2), g(4)g(4) and g(6)g(6).

g(2)g(2): the trapezoid from 0 to 2, 1+32⋅2=4\frac{1 + 3}{2} \cdot 2 = 4.

g(4)g(4): add the rectangle from 2 to 4, 3⋅2=63 \cdot 2 = 6. So g(4)=10g(4) = 10.

g(6)g(6): add the triangle from 4 to 6, 12(2)(3)=3\frac{1}{2}(2)(3) = 3. So g(6)=13g(6) = 13.

How fast does the area grow?

Top: the graph of f(t) = 0.5t + 1 on [0, 6], with the area under it from 0 to x = 3 shaded; that area is g(3) = 5.25 and the height f(3) is 2.5. Bottom: the accumulation function g(x) = x squared over 4 + x, with the point (3, 5.25) and a dashed tangent line there of slope g prime of 3 = f(3) = 2.5.
Figure 1. The area under f up to x is g(x); the slope of g at x is the height f(x). LevlPrep original diagram.

Move xx to the right by a tiny amount hh. The area gg gains a thin strip: width hh, height about f(x)f(x). So

g(x+h)−g(x)≈f(x)⋅h,sog(x+h)−g(x)h≈f(x).g(x + h) - g(x) \approx f(x)\cdot h, \quad\text{so}\quad \frac{g(x + h) - g(x)}{h} \approx f(x).

As h→0h \to 0 the approximation becomes exact, because ff is continuous. The left side becomes the derivative g′(x)g'(x). That is the Fundamental Theorem of Calculus (part 1):

If ff is continuous, then ddx∫axf(t)dt\frac{d}{dx}\int_a^x f(t)\,dt=f(x)= f(x).

In words: the rate at which area accumulates is the height of the graph. The figure shows it for f(t)=0.5t+1f(t) = 0.5t + 1: at x=3x = 3 the height is 2.5, and the slope of gg there is 2.5.

A function FF with F′=fF' = f is called an antiderivative of ff. So the theorem says: every accumulation function of a continuous ff is an antiderivative of ff. Even a function with no formula for its area, like e−t2e^{-t^2}, has antiderivatives: ∫0xe−t2dt\int_0^x e^{-t^2}\,dt is one.

The lower limit aa does not appear in the answer. Changing it changes where gg is 0, which shifts the whole graph of gg up or down, but not its slope. Writing f(x)−f(a)f(x) - f(a) is a common mistake.

Reading derivatives of g from the graph of f

  • g(x)g(x) is an area under ff.
  • g′(x)=f(x)g'(x) = f(x): read the height of ff.
  • g″(x)=f′(x)g''(x) = f'(x): read the slope of ff. So gg is concave up where ff is increasing and concave down where ff is decreasing.

With the graph in Worked example 1: g′(1)=f(1)=2g'(1) = f(1) = 2, and g″(3)=f′(3)=0g''(3) = f'(3) = 0 because ff is flat on (2,4)(2, 4).

A function as the upper limit: add the chain rule

If the upper limit is u(x)u(x) instead of xx, then ∫au(x)f(t)dt\int_a^{u(x)} f(t)\,dt is g(u(x))g(u(x)), a composition. The chain rule gives

ddx∫au(x)f(t)dt=g′(u(x))u′(x)=f(u(x))u′(x).\frac{d}{dx}\int_a^{u(x)} f(t)\,dt = g'(u(x))\,u'(x) = f(u(x))\,u'(x).

Worked example 2. Find h′(x)h'(x) for h(x)h(x)=∫0x3e−t2dt= \int_0^{x^3} e^{-t^2}\,dt.

Integrand: f(t)=e−t2f(t) = e^{-t^2}. Upper limit: u=x3u = x^3, so u′=3x2u' = 3x^2.

Evaluate the integrand at the upper limit: f(x3)f(x^3)=e−(x3)2= e^{-(x^3)^2}=e−x6= e^{-x^6}.

Multiply by u′u': h′(x)=3x2e−x6h'(x) = 3x^2 e^{-x^6}.

Two things are easy to miss: the factor 3x23x^2, and putting x3x^3 (not xx) into the integrand.

Accumulation with a starting amount

In context, an amount often equals a starting value plus an accumulation: W(t)W(t)=200= 200+∫0tr(s)ds+ \int_0^t r(s)\,ds liters of water. Then W(0)=200W(0) = 200, and by the theorem W′(t)=r(t)W'(t) = r(t): the amount changes at the given rate, exactly as it should.

The theorem has a second part, about evaluating ∫abf(x)dx\int_a^b f(x)\,dx with an antiderivative. It comes in topic 6.7, after you have learned more about accumulation functions and the properties of integrals.

Spot a mistake on this page?