Singularities of Analytic Complex Functions
Like in elementary calculus, it is important to study the behaviour of singularities of functions to obtain a better understanding of the function itself. We begin by giving a definition of a singularity for an analytic complex function.
Definition: A Singularity of an analytic function $f$ is a point $z_0$ for which $f$ is not analytic at $z_0$. A singularity $z_0$ of $f$ is said to be an Isolated Singularity if there exists an open disk $D(z_0, r)$ for which $f$ is analytic on the punctured disk $D(z_0, r) \setminus \{ z_0 \}$. |
For example, consider the function $\displaystyle{f(z) = \frac{z^3}{z^2 - 1}}$. Then $f$ is analytic everywhere it is defined but not analytic at points where $f$ is undefined. At $z = 1, -1$ $f$ is undefined and hence $f$ is not analytic at $z = 1, -1$. Therefore $1$ and $-1$ are singularities of $f$ - both of which are isolated as illustrated in the image below.

We will now look at the three main types of isolated singularities of complex analytic functions.
Definition: $z_0$ is a Removable Singularity of $f$ if $\displaystyle{\lim_{z \to z_0} f(z)}$ exists. |
For example, consider the function $\displaystyle{f(z) = \frac{z^2}{z}}$. The point $z_0 = 0$ is an isolated singularity of $f$ and notice that:
(1)Therefore $z_0 = 0$ is a removable singularity of $f$.
Definition: $z_0$ is a Pole Singularity of $f$ if $\displaystyle{\lim_{z \to z_0} \mid f(z) \mid = \infty}$. |
For example, consider the function $\displaystyle{g(z) = \frac{1}{z}}$. The point $z_0 = 0$ is an isolated singularity of $g$ and notice that:
(2)Therefore $z_0 = 0$ is a pole singularity of $g$.
Definition: $z_0$ is an Essential Singularity of $f$ if $\displaystyle{\lim_{z \to z_0} \mid f(z) \mid \neq \infty}$ and does not exist. |
For example, consider the function $\displaystyle{h(z) = e^{1/z}}$. The point $z_0 = 0$ is an isolated singularity of $h$ and notice that:
(3)To prove this, consider the limit of $h$ as $z \to 0$ along the positive $x$-axis:
(4)And consider the limit of $h$ as $z \to 0$ along the negative $x$-axis:
(5)Since these limits are not equal, we conclude that $\displaystyle{\lim_{z \to 0} \mid h(z) \mid}$ does not exist (and does not equal $\infty$.