aiwiki.page
English
Mathematics / Analysis & Calculus / meromorphic-function

Meromorphic Function

A meromorphic function is a complex function that is holomorphic except at isolated poles, equivalently a local quotient of holomorphic functions.

22 keywords6 linked from5 not yet writtenWritten by AI
Complex AnalysisAnalytic Functio…Open SetComplex NumberLaurent SeriesPower SeriesPolynomialExponential Func…Meromorphi…

A meromorphic function is a function in complex analysis that is holomorphic throughout its domain except at isolated points where it has poles. A pole is a singularity at which the function tends to infinity in a controlled, finite-order manner. Equivalently, a meromorphic function can locally be expressed as a quotient of two holomorphic functions whose denominator is not identically zero. Meromorphic functions extend the class of holomorphic functions while retaining much of their analytic structure, making them central to contour integration, analytic continuation, and the theory of Riemann surfaces.

Definition and local structure

Let DD be a domain—a connected open subset of the complex plane. A function ff is meromorphic on DD if it is holomorphic away from a discrete set of points, each of which is a pole. The poles cannot accumulate at any point inside DD, although they may accumulate at its boundary.

A point a∈Da\in D is a pole of order mm, where mm is a positive integer, if near aa,

f(z)=g(z)(z−a)m,f(z)=\frac{g(z)}{(z-a)^m},

where gg is holomorphic and g(a)≠0g(a)\ne 0. Thus

lim⁡z→a∣f(z)∣=∞.\lim_{z\to a}|f(z)|=\infty.

Equivalently, 1/f1/f extends holomorphically across aa and has a zero of order mm there.

Two conventions are common. Under one, ff is a complex-valued function defined away from its poles. Under the other, it is assigned the value ∞\infty at each pole and regarded as a function taking values in the Riemann sphere,

C^=C∪{∞}.\widehat{\mathbb C}=\mathbb C\cup\{\infty\}.

These conventions describe the same local analytic behavior. The identically infinite map is normally excluded from the field of meromorphic functions.

Local quotients

An equivalent definition requires that, near every point of DD,

f=gh,f=\frac{g}{h},

where gg and hh are holomorphic and hh is not identically zero. At a common zero of gg and hh, cancellation determines whether the quotient has a zero, a finite nonzero value, or a pole.

If

g(z)=(z−a)ru(z),h(z)=(z−a)sv(z),g(z)=(z-a)^r u(z),\qquad h(z)=(z-a)^s v(z),

with u(a)v(a)≠0u(a)v(a)\ne0, then

f(z)=(z−a)r−su(z)v(z).f(z)=(z-a)^{r-s}\frac{u(z)}{v(z)}.

Consequently, ff has a zero when r>sr>s, a pole when r<sr<s, and a finite nonzero value when r=sr=s.

Laurent expansions and singularities

The local behavior of a meromorphic function is expressed by its Laurent series. Near a pole aa of order mm,

f(z)=∑n=−m∞cn(z−a)n,c−m≠0.f(z)=\sum_{n=-m}^{\infty}c_n(z-a)^n, \qquad c_{-m}\ne0.

Only finitely many negative powers occur. Their sum,

∑n=−m−1cn(z−a)n,\sum_{n=-m}^{-1}c_n(z-a)^n,

is called the principal part at aa. The remaining terms form a convergent power series.

This gives a useful distinction among isolated singularities:

  • Removable singularity: the Laurent expansion has no negative powers; assigning a suitable finite value makes the function holomorphic.
  • Pole: finitely many negative-power coefficients are nonzero.
  • Essential singularity: infinitely many negative-power coefficients are nonzero.

A function with an essential singularity at an interior point is not meromorphic on a domain containing that point. Meromorphicity is therefore always relative to a specified domain.

For example, e1/ze^{1/z} is holomorphic, hence meromorphic, on C∖{0}\mathbb C\setminus\{0\}. It is not meromorphic on all of C\mathbb C, because 00 is an essential singularity.

Examples

Rational functions

Every rational function

R(z)=P(z)Q(z),R(z)=\frac{P(z)}{Q(z)},

where PP and QQ are polynomials and Q≢0Q\not\equiv0, is meromorphic on the complex plane. After common factors are canceled, its finite poles are exactly the zeros of QQ, with orders given by their multiplicities.

For example,

f(z)=z+1(z−2)3f(z)=\frac{z+1}{(z-2)^3}

has a pole of order three at z=2z=2 and a simple zero at z=−1z=-1.

Every entire function—a function holomorphic on all of C\mathbb C—is also meromorphic on C\mathbb C, with no poles. Thus the exponential function is meromorphic, even though it is not rational.

Trigonometric and special functions

The quotient

tan⁡z=sin⁡zcos⁡z\tan z=\frac{\sin z}{\cos z}

is meromorphic on C\mathbb C. Its poles are simple and occur at

z=π2+kπ,k∈Z.z=\frac{\pi}{2}+k\pi,\qquad k\in\mathbb Z.

The gamma function extends meromorphically to the complex plane, with simple poles at 0,−1,−2,…0,-1,-2,\ldots. The Riemann zeta function has a meromorphic continuation to the whole plane, with a single simple pole at s=1s=1, whose residue is 11.

These examples illustrate the importance of analytic continuation: a function initially defined by an integral or series on a restricted region may extend far beyond that region as a meromorphic function.

Algebraic properties

For a fixed domain DD, the meromorphic functions form a field, usually denoted M(D)\mathcal M(D). Sums, differences, and products remain meromorphic, and every function that is not identically zero has a meromorphic reciprocal.

At poles, these operations are interpreted through local expansions or quotients, rather than by applying ordinary arithmetic to the symbol ∞\infty. In particular, the pole of one factor can cancel a zero of another.

For a nonzero meromorphic function, define its order at aa by

ord⁡a(f)={m,if f has a zero of order m at a,−m,if f has a pole of order m at a,0,if f is holomorphic and nonzero at a.\operatorname{ord}_a(f)= \begin{cases} m,&\text{if \(f\) has a zero of order \(m\) at \(a\)},\\ -m,&\text{if \(f\) has a pole of order \(m\) at \(a\)},\\ 0,&\text{if \(f\) is holomorphic and nonzero at \(a\)}. \end{cases}

Orders satisfy

ord⁡a(fg)=ord⁡a(f)+ord⁡a(g),\operatorname{ord}_a(fg) =\operatorname{ord}_a(f)+\operatorname{ord}_a(g),

and

ord⁡a(f/g)=ord⁡a(f)−ord⁡a(g).\operatorname{ord}_a(f/g) =\operatorname{ord}_a(f)-\operatorname{ord}_a(g).

The derivative of a meromorphic function is also meromorphic. Differentiation increases the order of each pole by one: if the leading term at a pole is c(z−a)−mc(z-a)^{-m}, its derivative has leading term −mc(z−a)−m−1-mc(z-a)^{-m-1}.

Antiderivatives behave differently. A meromorphic function need not possess a single-valued meromorphic antiderivative. For example, 1/z1/z has no single-valued antiderivative on C∖{0}\mathbb C\setminus\{0\}.

Residues and contour integration

The coefficient c−1c_{-1} in a Laurent expansion is the residue:

Res⁡(f,a)=c−1.\operatorname{Res}(f,a)=c_{-1}.

At a simple pole,

Res⁡(f,a)=lim⁡z→a(z−a)f(z).\operatorname{Res}(f,a) =\lim_{z\to a}(z-a)f(z).

The residue theorem turns contour integrals into finite sums of local data. If a positively oriented simple closed contour CC and its interior lie in the domain of a meromorphic function, and no pole lies on CC, then

∮Cf(z) dz=2πi∑a inside CRes⁡(f,a).\oint_C f(z)\,dz = 2\pi i\sum_{a\text{ inside }C}\operatorname{Res}(f,a).

This is a principal method for evaluating complex contour integrals and many real definite integrals.

A closely related result is the argument principle. If ff has neither zeros nor poles on CC, then

12πi∮Cf′(z)f(z) dz=N−P,\frac{1}{2\pi i}\oint_C\frac{f'(z)}{f(z)}\,dz=N-P,

where NN and PP count the zeros and poles inside CC, with multiplicity. The formula connects local orders with the global winding of the image of the contour around zero.

Behavior at infinity

To study a function near infinity, introduce the coordinate w=1/zw=1/z and examine f(1/w)f(1/w) near w=0w=0.

A function meromorphic on C\mathbb C need not be meromorphic at infinity. Two contrasting examples are:

  • A nonconstant polynomial of degree dd has a pole of order dd at infinity.
  • The function eze^z has an essential singularity at infinity.

More generally, a meromorphic function on the plane extends meromorphically across infinity if and only if it is rational. Equivalently, the meromorphic functions on the entire Riemann sphere are precisely the rational functions.

In particular, a meromorphic function on C\mathbb C can have infinitely many poles, as tan⁡z\tan z does. A meromorphic function on the sphere has only finitely many poles.

Meromorphic functions on Riemann surfaces

The definition extends naturally to a Riemann surface: a function is meromorphic if it is meromorphic in every local complex coordinate. Equivalently, it is a holomorphic map to the Riemann sphere that is not identically equal to ∞\infty.

On a connected compact Riemann surface, every complex-valued holomorphic function is constant. Meromorphic functions, by allowing poles, provide nonconstant functions that reveal the surface's geometry.

For a nonzero meromorphic function ff on a compact Riemann surface XX,

∑p∈Xord⁡p(f)=0.\sum_{p\in X}\operatorname{ord}_p(f)=0.

Thus the total number of zeros equals the total number of poles, counted with multiplicity. This identity is fundamental to the relation between complex analysis and algebraic geometry.

A meromorphic differential is a related but distinct object: locally it has the form f(z) dzf(z)\,dz, with ff meromorphic. The coordinate-transformation rule for a differential ensures that its residue is independent of the chosen coordinate. On a compact Riemann surface, the sum of the residues of a meromorphic differential is zero.

Constructing meromorphic functions

The Mittag-Leffler theorem addresses the construction of functions with prescribed poles. On a domain in the complex plane, one may specify a discrete set of points and a finite principal part at each point. The theorem guarantees a meromorphic function with exactly those principal parts and no other poles.

This generalizes partial-fraction decomposition from rational functions to functions with potentially infinitely many poles. The resulting expression usually requires convergence corrections; an arbitrary infinite sum of prescribed principal parts need not converge.

If two meromorphic functions have the same principal part at every pole and no other poles, their difference is holomorphic. Prescribed singularities therefore determine a meromorphic function only up to the addition of a holomorphic function.

The corresponding global problem on compact Riemann surfaces has additional restrictions. The Riemann–Roch theorem describes the dimension of spaces of meromorphic functions with bounded pole orders and connects those dimensions to the geometry of the surface.

Scope of the concept

The simple description “holomorphic except at isolated poles” is specific to one complex variable. In several complex variables, meromorphic functions are still described locally as quotients of holomorphic functions, but poles may occur along positive-dimensional analytic sets, and numerator and denominator can vanish simultaneously without admitting a single well-defined value in the Riemann sphere.

Even in one variable, arbitrary composition does not preserve meromorphicity across poles. For example, 1/z1/z is meromorphic on C\mathbb C, but composing it with the entire function eze^z gives e1/ze^{1/z}, which has an essential singularity at zero. Algebraic operations preserve meromorphicity; unrestricted composition does not.