ConceptComplete

Meromorphic Functions

Meromorphic functions are the natural generalization of rational functions: they are holomorphic except for isolated poles. The theory of meromorphic functions connects to algebraic geometry through the study of divisors on Riemann surfaces.


Definition and Examples

Definition9.3Meromorphic function

A function ff on a domain DD is meromorphic if it is holomorphic on DD except for a set of isolated points, each of which is a pole. Equivalently, ff can be locally expressed as a ratio g/hg/h of holomorphic functions with h≑̸0h \not\equiv 0.

ExampleExamples of meromorphic functions
  • Rational functions P(z)/Q(z)P(z)/Q(z) are meromorphic on C\mathbb{C} (and on C^\hat{\mathbb{C}}).
  • tan⁑z=sin⁑z/cos⁑z\tan z = \sin z / \cos z is meromorphic on C\mathbb{C} with poles at z=(n+1/2)Ο€z = (n + 1/2)\pi.
  • Ξ“(z)\Gamma(z) is meromorphic on C\mathbb{C} with simple poles at z=0,βˆ’1,βˆ’2,…z = 0, -1, -2, \ldots
  • The Weierstrass β„˜\wp-function is meromorphic on C\mathbb{C} with double poles at lattice points.

Mittag-Leffler Theorem

Theorem9.3Mittag-Leffler theorem

Given a sequence of distinct points {an}βŠ‚C\{a_n\} \subset \mathbb{C} with ∣anβˆ£β†’βˆž|a_n| \to \infty and principal parts (Laurent tails) pn(z)=βˆ‘k=1mncn,k/(zβˆ’an)kp_n(z) = \sum_{k=1}^{m_n} c_{n,k}/(z-a_n)^k, there exists a meromorphic function ff on C\mathbb{C} whose only singularities are poles at {an}\{a_n\} with the prescribed principal parts.

RemarkNon-uniqueness

The function ff is not unique: if f1,f2f_1, f_2 are two solutions, then f1βˆ’f2f_1 - f_2 is entire. The Mittag-Leffler theorem is the additive analogue of the Weierstrass factorization theorem (which is multiplicative).

ExamplePartial fraction of $\cot(\pi z)$

The function Ο€cot⁑(Ο€z)\pi\cot(\pi z) has simple poles at all integers with residue 11. By the Mittag-Leffler theorem:

Ο€cot⁑(Ο€z)=1z+βˆ‘n=1∞(1zβˆ’n+1z+n)=1z+2zβˆ‘n=1∞1z2βˆ’n2.\pi\cot(\pi z) = \frac{1}{z} + \sum_{n=1}^{\infty}\left(\frac{1}{z-n} + \frac{1}{z+n}\right) = \frac{1}{z} + 2z\sum_{n=1}^{\infty}\frac{1}{z^2-n^2}.

Setting z=0z = 0 in the derivative yields the evaluation βˆ‘n=1∞1/n2=Ο€2/6\sum_{n=1}^\infty 1/n^2 = \pi^2/6.


Meromorphic Functions on the Riemann Sphere

Definition9.4Meromorphic on $\hat{\mathbb{C}}$

A meromorphic function on the Riemann sphere C^=Cβˆͺ{∞}\hat{\mathbb{C}} = \mathbb{C} \cup \{\infty\} is one that is meromorphic on C\mathbb{C} and also meromorphic at ∞\infty (when expressed in the local coordinate w=1/zw = 1/z). The meromorphic functions on C^\hat{\mathbb{C}} are precisely the rational functions.

Theorem9.4Characterization of rational functions

ff is meromorphic on C^\hat{\mathbb{C}} if and only if f=P(z)/Q(z)f = P(z)/Q(z) for polynomials P,QP, Q. This is because a meromorphic function on a compact Riemann surface has finitely many poles, and the principal parts and polynomial growth at ∞\infty force ff to be rational.


Elliptic Functions

Definition9.5Elliptic function

An elliptic function is a meromorphic function on C\mathbb{C} that is doubly periodic: f(z+Ο‰1)=f(z+Ο‰2)=f(z)f(z + \omega_1) = f(z + \omega_2) = f(z) for two R\mathbb{R}-linearly independent periods Ο‰1,Ο‰2\omega_1, \omega_2.

RemarkProperties of elliptic functions

Every non-constant elliptic function:

  1. Must have poles (by Liouville's theorem, a holomorphic doubly periodic function is constant).
  2. Has at least two poles (counted with multiplicity) in each fundamental parallelogram.
  3. Satisfies: (sum of residues in a period parallelogram) = 00.
  4. Takes every value in C^\hat{\mathbb{C}} the same number of times (the order of the function).
ExampleThe Weierstrass $\wp$-function

For a lattice Ξ›=ZΟ‰1+ZΟ‰2\Lambda = \mathbb{Z}\omega_1 + \mathbb{Z}\omega_2, the Weierstrass β„˜\wp-function is

β„˜(z)=1z2+βˆ‘Ο‰βˆˆΞ›βˆ–{0}(1(zβˆ’Ο‰)2βˆ’1Ο‰2).\wp(z) = \frac{1}{z^2} + \sum_{\omega \in \Lambda \setminus \{0\}}\left(\frac{1}{(z-\omega)^2} - \frac{1}{\omega^2}\right).

It is elliptic of order 22 and satisfies the differential equation (β„˜β€²)2=4β„˜3βˆ’g2β„˜βˆ’g3(\wp')^2 = 4\wp^3 - g_2\wp - g_3 where g2,g3g_2, g_3 are the Eisenstein invariants of Ξ›\Lambda.