Integration and series · Definition

Integrability of continuous functions

real analysisintegration

Before you start you need to know

It tells us a broad class of functions can be integrated without building Darboux sums by hand, and it underlies computing integrals via antiderivatives.

Definition

Result stating that every function \(f\) continuous on a closed, bounded interval \([a,b]\) is Riemann integrable there.

\[f\text{ continuous on }[a,b]\ \Rightarrow\ \int_a^b f(x)\,dx\text{ exists}\]

The proof idea is Riemann's criterion: one partition with \(U(f,P)-L(f,P)<\varepsilon\) suffices. Uniform continuity gives a \(\delta\) that works on the whole interval; if every subinterval is shorter than \(\delta\), the oscillation of \(f\) on it is at most \(\varepsilon/(b-a)\), and summing the area differences gives a gap of at most \(\varepsilon\).

The hypotheses matter. A closed, bounded interval is what turns continuity into uniform continuity; on \((0,1]\), \(1/x\) is continuous but unbounded and not integrable. Continuity is sufficient, not necessary: a bounded function with a jump, or finitely many discontinuities, is still integrable.

Running example: \(f(x)=x^2\) on \([0,1]\) is continuous, so its integral exists immediately; computing it with sums gives \(\tfrac13\). The converse fails: the Dirichlet function is nowhere continuous and not integrable, but other discontinuous functions are integrable.

This result guarantees that \(\int_a^x f\) makes sense for every continuous \(f\) and every \(x\in[a,b]\), which is needed for the fundamental theorem of calculus, now relating that integral to the derivative.

Think about it

Why is \(f(x)=1/x\) on \((0,1]\) continuous but not Riemann integrable?

The interval is not closed and \(f\) is unbounded: suprema on subintervals near \(0\) are arbitrarily large, so there are no finite upper sums.

If \(f\) is integrable on \([a,b]\), must it be continuous?

No. A function with a jump, such as \(0\) up to \(\tfrac12\) and \(1\) afterwards, is integrable and not continuous.

Exercises 10

References

  • Stephen Abbott, Understanding Analysis, 2nd edition, Springer, 2015. §7.2, Teorema 7.2.9
  • Walter Rudin, Principles of Mathematical Analysis, 3rd edition, McGraw-Hill, 1976. Cap. 6, Teorema 6.8
  • Robert G. Bartle, Donald R. Sherbert, Introduction to Real Analysis, 4th edition, Wiley, 2011. §7.2