# Riemann integral

A generalization of the concept of a Cauchy integral to a certain class of discontinuous functions; introduced by B. Riemann (1853). Consider a function $f$ which is given on an interval $[a,b]$. Let $a=x_0<x_1<\dots<x_n=b$ is a partition (subdivision) of the interval $[a,b]$ and $\Delta x_i = x_i-x_{i-1}$, where $i=1,\dots,n$. The sum
\begin{equation}\tag{1}
\sigma = f(\xi_1)\Delta x_1+\dots+f(\xi x_i)\Delta x_i +\dots +f(x_n)\Delta x_n,
\end{equation}
where $x_{i-1}\leq\xi_i\leq x_i$, is called the Riemann sum corresponding to the given partition of $[a,b]$ by the points $x_i$ and to the sample of points $\xi_i$. The number $I$ is called the limit of the Riemann sums (1) as $\max_i \Delta x_i \to 0$ if for any $\varepsilon>0$ a $\delta>0$ can be found such that $\max_i \Delta x_i < \delta$ implies the inequality $|\sigma - I|<\varepsilon$. If the Riemann sums have a finite limit $I$ as $\max_i \Delta x_i \to 0$, then the function $f$ is called Riemann integrable over $[a,b]$, where $a< b$. The limit is known as the definite Riemann integral of $f$ over $[a,b]$, and is written as
\begin{equation}\tag{2}
\int\limits_a^bf(x)\,dx.
\end{equation}
When $a=b$ then, by definition,
\begin{equation}
\int\limits_a^af(x)\,dx = 0,
\end{equation}
and when $a>b$ the integral (2) is defined using the equation
\begin{equation}
\int\limits_a^bf(x)\,dx = -\int\limits_b^af(x)\,dx.
\end{equation}
A necessary and sufficient condition for the Riemann integrability of $f$ over $[a,b]$ is the boundedness of $f$ on this interval and the zero value of the Lebesgue measure of the set of all points of discontinuity of $f$ contained in $[a,b]$.

## Properties of the Riemann integral

- Every Riemann-integrable function $f$ on $[a,b]$ is also bounded on this interval (the converse is not true: The Dirichlet function is an example of a bounded and non-integrable function on $[a,b]$).
- The linearity property: For any constants $\alpha$ and $\beta$, the integrability over $[a,b]$ of both functions $f$ and $g$ implies that the function $\alpha f + \beta g$ is integrable over this interval, and the equation \begin{equation} \int\limits_a^b[\alpha f(x) + \beta g(x)]\,dx = \alpha\int\limits_a^bf(x)\,dx + \beta\int\limits_a^bg(x)\,dx \end{equation} holds.
- The integrability over $[a,b]$ of both functions $f$ and $g$ implies that their product $fg$ is integrable over this interval.
- Additivity: The integrability of a function $f$ over both intervals $[a,c]$ and $[c,b]$ implies that $f$ is integrable over $[a,b]$, and \begin{equation} \int\limits_a^bf(x)\,dx = \int\limits_a^cf(x)\,dx + \int\limits_c^bf(x)\,dx. \end{equation}
- If two functions $f$ and $g$ are integrable over $[a,b]$ and if $f(x)\geqslant g(x)$ for every $x$ in this interval, then \begin{equation} \int\limits_a^bf(x)\,dx \geqslant \int\limits_a^bg(x)\,dx. \end{equation}
- The integrability of a function $f$ over $[a,b]$ implies that the function $|f|$ is integrable over this interval, and the estimate \begin{equation} \left|\int\limits_a^bf(x)\,dx\right| \leqslant \int\limits_a^b|f(x)|\,dx \end{equation} holds.
- The mean-value formula: If two real-valued functions $f$ and $g$ are integrable over $[a,b]$, if the function $g$ is non-negative or non-positive everywhere on this interval, and if $M$ and $m$ are the least upper and greatest lower bounds of $f$ on $[a,b]$, then a number $\mu$ can be found, $m\leqslant\mu\leqslant M$, such that the formula \begin{equation}\tag{3} \int\limits_a^bf(x)g(x)\,dx = \mu\int\limits_a^bg(x)\,dx, \end{equation} holds. If, in addition, $f$ is continuous on $[a,b]$, then this interval will contain a point $\xi$ such that in formula (3), \begin{equation} \mu = f(\xi). \end{equation}
- The second mean-value formula (Bonnet's formula): If a function $f$ is real-valued and integrable over $[a,b]$ and if a function $g$ is real-valued and monotone on this interval, then a point $\xi$ can be found in $[a,b]$ such that the formula \begin{equation} \int\limits_a^bf(x)g(x)\,dx = g(a)\int\limits_a^{\xi}f(x)\,dx + g(b)\int\limits_{\xi}^bf(x)\,dx, \end{equation} holds.

#### References

[1] | B. Riemann, "Ueber die Darstellbarkeit einer Function durch eine trigonometrische Reihe" H. Weber (ed.) , B. Riemann's Gesammelte Mathematische Werke , Dover, reprint (1953) pp. 227–271 ((Original: Göttinger Akad. Abh. 13 (1868))) |

[2] | V.A. Il'in, E.G. Poznyak, "Fundamentals of mathematical analysis" , 1–2 , MIR (1982) (Translated from Russian) |

[3] | L.D. Kudryavtsev, "A course in mathematical analysis" , 1–2 , Moscow (1988) (In Russian) |

[4] | S.M. Nikol'skii, "A course of mathematical analysis" , 1–2 , MIR (1977) (Translated from Russian) |

[a1] | G.E. Shilov, "Mathematical analysis" , 1–2 , M.I.T. (1974) (Translated from Russian) |

[a2] | I.N. Pesin, "Classical and modern integration theories" , Acad. Press (1970) (Translated from Russian) |

[a3] | K.R. Stromberg, "Introduction to classical real analysis" , Wadsworth (1981) |

[a4] | W. Rudin, "Principles of mathematical analysis" , McGraw-Hill (1976) pp. 75–78 |

**How to Cite This Entry:**

Riemann integral.

*Encyclopedia of Mathematics.*URL: http://www.encyclopediaofmath.org/index.php?title=Riemann_integral&oldid=38637