Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If we have an analytic function $f(z)$ in the upper half-plane, which is also continuous on the real line. If we define a new function as $$ F(z) = \left\{ \begin{array}{lr} f(z) &: z\in \mathbb U \cup \mathbb R\\ f^{\#}(z) & :z\in \mathbb L \;\;\;\;\;\;\; \end{array} \right. $$ where $\mathbb U $ is the open half-plane, and $\mathbb L $is the open lower half plane, $f^{\#}(z)=\bar{f}(\bar{z})$. Is the function $F(z)$ an entire function? Why?

share|cite|improve this question
You may want the additional hythesis that the restriction of $f$ to the real line is real-valued. Also, if this is homework, please tell it, and explain what you have done; otherwise, I'd like to know the motivation. – D. Thomine Jun 14 '12 at 0:47
You can use Morera's theorem. – Jonas Meyer Jun 14 '12 at 0:48
@D.Thomine: You mean the problem could be in the $\#$ sign? If this is the case I think we can use the definition $$ F(z) = \left\{ \begin{array}{lr} f(z) &: z\in \mathbb U \cup \mathbb R\\ f(\bar{z}) & :z\in \mathbb L \;\;\;\;\;\;\; \end{array} \right. $$ instead. (BTW, this is not a homework) – Mate Jun 14 '12 at 0:51
The new definition is very different, and I'm pretty sure isn't in general entire (whereas your first one was, if you add the real-on-reals condition). – Ben Millwood Jun 14 '12 at 0:55
@benmachine: No I don't have this information, $f$ is not necessary real-on-reals. – Mate Jun 14 '12 at 1:00

To summarise the conclusions of the comments:

  • If $f$ isn't real-valued on $\mathbb R$, say $f(a) = b + ci$ for some $a,b,c\in \mathbb R$, $c \not= 0$, then we have $$\begin{align} \lim_{n\to\infty} F(a+i/n)&=\lim_{n\to\infty}f(a+i/n)\\ &= f(a) = b + ci\\ \lim_{n\to\infty} F(a-i/n)&=\;\lim_{n\to\infty}\overline{f(\overline{a-i/n})} \\ &=\lim_{n\to\infty} \overline{f(a+i/n)} \\ &= \overline{f(a)} = b - ci \end{align}$$. Hence the limits from above the real axis and below disagree at $a$, so $F$ is not continuous, hence cannot possibly be analytic.
  • If $f$ is real-valued on $\mathbb R$ then Morera's theorem will give that $F$ is analytic (hence entire): this result is called the Schwarz reflection principle.
share|cite|improve this answer
If I understand the argument well, then the function $F$ defined by $$ F(z) = \left\{ \begin{array}{lr} f(z) &: z\in \mathbb U \cup \mathbb R\\ g(z) & :z\in \mathbb L \;\;\;\;\;\;\; \end{array} \right. $$ is entire if $f$ is analytic in $\mathbb U\cup\mathbb R$, $g$ is analytic in $\mathbb L\cup\mathbb R$, and $f(x)=g(x)$ for all $x\in \mathbb R$, I'm right!! – Mate Jun 14 '12 at 1:54
@Mate: I think that's correct, but note that for each $f$ there will (unless I'm mistaken) only be at most one $g$ that satisfies those conditions. – Ben Millwood Jun 14 '12 at 11:40
I didn't get it? What I have is that, given an analytic functions $f$ in the closed upper half plane and another analytic function $g$ in the closed lower half plane which agrees with $f$ on the real axis, then I ddefine a function $F$ asin my comment above, and I think it must be entire. Is this true? – Mate Jun 14 '12 at 16:06
@Mate: yes. My comment only meant that if you're given $f$, there might not be an analytic $g$ that is equal to it on the real axis, and if there is such a $g$, then there's only one such $g$. – Ben Millwood Jun 14 '12 at 18:18
Why there should be only one such $g$? – Mate Jun 14 '12 at 20:23

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.