Build the definite integral from finite sums, interpret its units and sign, and connect accumulation to antiderivatives.
By Theory Commons Editors4 min readPublished Aug 22, 2026Updated Aug 23, 2026
Suppose a flow meter reports liters per second. One reading tells us a rate, not the amount of water that has passed. To recover volume, we must combine the rate over time. If the rate changes continuously, ordinary multiplication is only an approximation. The definite integral is the limit of increasingly accurate finite sums.
Learning objectives
By the end of this lesson, you should be able to:
construct a Riemann sum from a partition;
explain the roles of f(xi∗) and Δxi;
distinguish signed accumulation from geometric area;
predict the units of an integral;
use both parts of the Fundamental Theorem of Calculus;
distinguish a definite integral from an indefinite integral.
Activate the prerequisite: finite accumulation
If velocity is constant at 6m/s for 4s, displacement is
(6m/s)(4s)=24m.
If velocity is 2m/s for two seconds and then 5m/s for three seconds, add the contributions:
(2)(2)+(5)(3)=19m.
This is already integration in finite form: rate times interval, summed across intervals.
From rectangles to a Riemann sum
Partition [a,b] at points
a=x0<x1<⋯<xn=b.
The ith width is Δxi=xi−xi−1. Choose a sample point xi∗ in that subinterval. If f changes little there, the contribution is approximately
f(xi∗)Δxi.
Adding every contribution gives the Riemann sum
i=1∑nf(xi∗)Δxi.
The sample value gives a local rate, height, or density. The width supplies the amount of input over which that value acts. Their product has the units of an accumulated quantity.
Worked construction: integrate f(x)=x
Divide [0,1] into n equal pieces. Then Δx=1/n. Using right endpoints, xi∗=i/n, so
Rn=i=1∑nnin1=n21i=1∑ni.
Using ∑i=1ni=n(n+1)/2,
Rn=2n2n(n+1)=21+2n1.
Therefore
∫01xdx=n→∞limRn=21.
The right sums overestimate because f(x)=x is increasing. Left sums underestimate. Both squeeze toward the same value.
Sign and geometric area
The integral is signed accumulation. Contributions above the horizontal axis are positive; those below are negative.
For f(x)=x on [−1,1],
∫−11xdx=0
because the negative and positive contributions cancel. The geometric area is not zero:
∫−11∣x∣dx=1.
Units expose meaning
If v(t) is in meters per second and dt is in seconds, then v(t)dt is in meters. If density ρ(x) is kilograms per meter and dx is meters, then ρ(x)dx is kilograms.
Treating dx as a reminder of the input variable and its units prevents many modeling errors.
The accumulation function
Define
A(x)=∫axf(t)dt.
If the upper endpoint increases by a small h, the new accumulated amount is approximately a thin rectangle:
A(x+h)−A(x)≈f(x)h.
Dividing by h suggests A′(x)≈f(x). The limiting statement is exact.
Worked application: displacement versus distance
Let velocity be v(t)=3t2−6t on 0≤t≤3. An antiderivative is V(t)=t3−3t2, so displacement is
∫03v(t)dt=V(3)−V(0)=0.
Zero displacement does not mean no motion. Since v(t)=3t(t−2) changes sign at t=2, total distance is
−∫02v(t)dt+∫23v(t)dt=4+4=8.
The sign split is a conceptual step, not merely algebra.
Definite and indefinite integrals
The definite integral ∫abf(x)dx is a number. The notation
∫f(x)dx=F(x)+C
denotes a family of antiderivatives. The constant C is required because differentiation erases constants. In F(b)−F(a), the same constant cancels.
Check your understanding
If f is negative everywhere on [a,b], what sign must ∫abf(x)dx have?
A flow rate is measured in gallons per minute. What are the units of its integral with respect to minutes?
Why can a definite integral be zero even when the integrand is not zero?
Show the reasoning
Negative: every sufficiently small contribution f(xi∗)Δxi is negative.
Gallons, because (gallons/minute)(minutes)=gallons.
Positive and negative contributions can cancel, as they do for an odd function over a symmetric interval.