The limits , and have the same denominator and need three different responses: the first is 2 after factoring, the second has no finite limit, and the third has different one-sided limits, and . No limit tells you its method. You find it by substituting first and then reading two things: the form substitution produced, and the structure of the expression.
Substitute first, then read the result
Substituting is always the first move, because it either finishes the problem or tells you what kind of problem you have.
After any rewrite, read the result of substituting again. A can become a number, another (rewrite again), or a nonzero number over 0 (no finite limit).
Matching a to its rewrite
The structure of the expression tells you which rewrite will expose a common factor.
| What you see | What to do | Example |
|---|---|---|
| Polynomials on top and bottom | Factor and cancel | |
| A square root in a sum or difference | Multiply by the conjugate | |
| Fractions inside a fraction | Combine over a common denominator | |
| A power that isn't expanded | Expand, then cancel | |
| Trig functions with a factorable identity | Rewrite with the identity | near |
| over a power of | Arrange and use its limit, 1 |
Sometimes two methods work. In you can multiply by the conjugate, or you can see as a difference of squares, , and cancel directly. Either is correct; the factoring is shorter.
The last row of the table belongs to topic 1.8, which proves . To use it, the angle inside the sine and the quantity in the denominator must match and both approach 0: .
When the formula changes: absolute values and pieces
An absolute value is a piecewise function in disguise: when , and when . At the formula switches, so split the limit into two sides and remove the absolute value on each:
The one-sided limits differ, so does not exist. A piecewise function works the same way, with the formulas given to you: each one-sided limit uses the formula for its own side, and each of those may need its own method.
When no rewrite works
Some limits can't be fixed by algebra, and recognizing them saves time.
- A bounded factor that oscillates, times a factor that approaches 0, such as at . The oscillating factor has no limit, so the product law fails, but the squeeze theorem (topic 1.8) gives 0.
- Only a graph or a table is given. Read or estimate the one-sided limits from it (topics 1.3 and 1.4); there is no formula to manipulate.
- Unit 4 adds a tool. L'Hospital's rule (topic 4.7) settles many limits using derivatives. Until then, the methods above are your tools, and they stay the faster route for most rational and radical expressions.
Worked examples
Find each limit, or explain why it does not exist.
(a) (b) (c) (d)
Show solution
(a) Substitution gives . The denominator isn't 0 and the function is rational, so the limit is 1.
(b) Substitution gives , and both are polynomials, so factor:
Substituting again gives , a nonzero number over 0, so there is no finite limit. Checking signs: for just left of , is a small negative number, so ; just right of the quotient goes to . The limit does not exist.
(c) Substitution gives . Factor for , then cancel:
(d) The absolute value switches formula at 0, so split. For , and the expression is . For , and the expression is . The right-hand limit is 2 and the left-hand limit is 0, so the limit does not exist.
Find (a) and (b) .
Show solution
Both are at , and neither has a factor you can cancel directly.
(a) The sine's argument is , so the known limit needs in the denominator. Multiply and divide by :
As , , so , and the product approaches .
(b) The suggests the identity , which needs a factor of . Multiply top and bottom by it:
The first factor approaches 1 and the second approaches , so the limit is .
Let be the function defined by
(a) Find .
(b) Find and .
(c) Does exist? Justify your answer.
Show solution
For , factor once and reuse it: for , .
(a) Near 0 the first formula applies, and substitution works: .
(b) Near the first formula applies, and gives at , a nonzero number over 0. The numerator is near 2. Just left of , is a small negative number, so ; just right, it is a small positive number, so .
(c) The formula changes at 3, so find each one-sided limit with its own formula.
From the left, .
From the right, substitution gives and there is a root in a difference, so multiply by the conjugate:
so .
Yes, the limit exists: , so . This is true even though is not defined.
Where students lose points
On the exam
Limit questions give you the expression and ask for the value, without naming a method, so the recognition step in this topic is part of every one of them. When a free-response part asks whether a limit exists, especially for a piecewise function like Example 3, the justification is both one-sided limits, each shown, and a statement that they are equal (or that they differ).
Check yourself
Question 1
Substituting into a rational function gives . Which statement is true?
- A
- B
- C
Factoring and canceling will reveal .
- D
is not a real number.
Show answer
D
A numerator near 3 divided by a denominator near 0 has unbounded size, so there is no finite limit. Choice C is the tempting one, but the denominator's factor isn't in the numerator (the numerator is 3 there), so nothing cancels.
- A
Question 2
Find .
Show answer
Factor the denominator as and rearrange: . The middle factor approaches 1 (topic 1.8) and the last approaches , giving .
Question 3
What is ?
- A
- B
- C
- D
Show answer
A
For , , so the expression is , which approaches . Choice B drops the sign the absolute value introduces on the left side.
- A
Question 4
For which function does substituting give directly?
- A
- B
- C
- D
Show answer
A
In A the denominator is at and every piece is continuous, so the limit is . B is (its limit, 1, takes topic 1.8), C changes formula at 0, and D is not defined at 0 and needs the squeeze theorem.
- A
Question 5
Find .
Show answer
Combine the top over the common denominator : the expression becomes for , which approaches .