Definition

Let and be metric spaces. Let and

Let be a limit point of ( need not be in ). Let . We say

if such that for all with

As we get closer to by , we get closer to by .

Example 1

Let where . Then is a limit point of but not in . However, we can claim .

Proof: Let . Choose . If and then

and so we are done.

Example 2

then does not exist.

Proof: Suppose by contradiction the limit is . Let . Then such that

By Archimedean Property, such that . By picking larger, we can get . Then , note and , so

which is a contradiction.

Lemma (Sequential Characterization of Limits)

Let be metric spaces. Let be a limit point, . Then

Proof: Suppose . Let . Then such that . We have .

We’ve chosen some , and rewrote the definition of function limits. The goal is that if and , , then . (Unfold definitions.)

Since such that , and since . Hence, by definition of , we have that . Then .

Suppose and that .

The goal is to show . We show by contradiction since the goal is static.

Suppose . Then

Apply for , this negation to . Then but

Now we have , since then and . So by assumption, . This is a contradiction.

Corollary (Uniqueness of Limits)

If has a limit at , then the limit is unique.

Proof: We apply the previous lemma and the uniqueness of of limits of Sequences.

Corollary (Algebra of Limits)

Let . Suppose

  1. Then
  2. If ,

Proof: This follows from Lemma (Sequential Characterization of Limits) and Lemma (Algebra of Sequences).

Lemma (Continuity = Limits)

If and is a limit point of , then

Proof: For every such that , if , then . But then this is the definition of limits as the limit .

Let . Choose as in the definition of limit. Suppose now that . We have 2 cases:

  1. . Then . Moreover,
  2. . Then . By definition of limits, .

Corollary (Sequential Characterization of Continuity)

Let , . Then is continuous at we have .

Corollary (Algebra of Continuous Functions)

Let where are metric spaces. Then if are continuous at , then

  1. is continuous at .
  2. is continuous at .
  3. is continuous at if