Calculus I › Limits › full formula sheet

limx→a [f(x) · g(x)] = L · M

Say it: “the limit as x approaches a of f of x times g of x equals L times M”

The product law for limits

The limit of a product is the product of the limits — when both pieces converge.

Notation on this page: L = limx→a f(x) and M = limx→a g(x) are assumed to exist (finite).

Before this lesson: The sum / difference law

Notation in this lesson

L·M
the product of the two individual limits
hypothesis
both factor limits must exist and be finite — or the law is silent

Where it comes from

The problem this law solves is the same divide-and-conquer as the sum law, one operation over: lim(x→4) x·√x should just be 4·2 = 8. And it is — provided both factors converge. The subtlety is that the converse fails, and failing to notice costs marks:

Before reading on: limx→4 x·√x — why is it legal to take the two limits separately here? What's being assumed?
lim(x→0) x·sin(1/x)
=
0
True, but not by the product law. lim(x→0) sin(1/x) does not exist (it oscillates between −1 and 1 forever), so the law’s hypothesis fails. The answer 0 needs the squeeze theorem (see its page).
“factor it”
=
lim(x→0) x · lim(x→0) sin(1/x)
Illegal. The second factor has no limit, so the product law cannot be invoked. A product’s limit can exist without each factor having one — the law goes one way only.

The intuition: if f(x) hugs L and g(x) hugs M, then f(x)·g(x) hugs L·M. The error analysis is a little richer than the sum law’s: f·g − L·M = f·(g−M) + M·(f−L). Each term pairs one factor’s error with the other factor’s size — and the proof below shows both terms vanish.

Derivation

Assume lim(x→a) f(x) = L and lim(x→a) g(x) = M, both finite. We prove lim(x→a) [f(x)·g(x)] = L·M.

fg − LM
=
f·(g − M) + M·(f − L)
Step 1 — the trick: add zero. Expand the right side: fg − fM + Mf − LM = fg − LM. We inserted −fM + Mf = 0, the same add-and-subtract move as the product rule proof. Now each group pairs one factor’s error with the other’s size.
|f(x)| ≤ B
for x
near a
Step 2 — f is bounded near a. Since f converges, it cannot escape to infinity: some B and some punctured neighborhood have |f(x)| ≤ B. (Take epsilon = 1 in the limit definition.) Without this, the term |f|·|g−M| could blow up.
|fg − LM|
≤
B·|g − M| + |M|·|f − L| < ε
Step 3 — split the budget. Choose δ so |g−M| < ε/(2B) and |f−L| < ε/(2(|M|+1)). Then the first term is below ε/2 and the second below ε/2. (The +1 avoids dividing by zero when M = 0.) ∎

Key steps shown; the argument above is complete. Notice the constant multiple law hiding inside: it is the product law with g(x) = c, whose limit is c. And the sum law’s min-delta trick appears again whenever two approximations must hold at once.

How to use it

The procedure, every time:

  1. Check the hypothesis. Do lim f and lim g both exist (finite)? If a factor oscillates or blows up, stop — reach for the squeeze theorem or simplify first.
  2. Evaluate each factor separately, then multiply the results.
  3. Three or more factors? Apply the law repeatedly: lim(f·g·h) = lim((f·g)·h).
  4. Watch for disguised products: (x−2)·(1/(x−2)) looks splittable but the second factor diverges — simplify to 1 first.

When the product law is the wrong tool

If one factor’s limit does not exist, the law is off the table — but the limit itself might still exist. The classic rescue is the squeeze theorem: a bounded oscillation times something going to 0 gives 0. Pattern to memorize: bounded × vanishing = 0.

Common mistake: writing lim(f·g) = lim f · lim g and plugging in “DNE” or infinity for a divergent factor, then treating the result as an answer. Check convergence first, always.

Worked examples

Four problems, easiest first. In each one, read every step — the why of each move is the lesson.

Example 1 — the basic move: lim(x→4) x·√x

  1. Check the hypothesis. lim(x→4) x = 4 and lim(x→4) √x = 2 both exist.
  2. Evaluate each factor. 4 and 2.
  3. Multiply. 4 · 2 = 8.
Common mistake: none — this is the law working as intended. The traps are all in the next examples.
Your turn: limx→9 x·√x = ?

Answer: 27.

Both factors converge: lim x = 9 and lim √x = 3, so 9·3 = 27.

Example 2 — with trigonometry: lim(x→π/2) x·sin x

  1. Check the hypothesis. lim(x→π/2) x = π/2; lim(x→π/2) sin x = 1 (sine is continuous). Both exist.
  2. Multiply. (π/2) · 1 = π/2.
  3. Why this is safe: sin x is not oscillating wildly here — near π/2 it hugs 1. Oscillation only breaks the law when the limit truly DNE.
Common mistake: evaluating sin at π/2 as 0 (confusing sine and cosine values). sin(π/2) = 1, cos(π/2) = 0.
Your turn: limx→π x·cos x = ?

Answer: −π.

lim x = π; lim cos x = cos π = −1 (cosine is continuous). Product: −π.

Example 3 — three factors: lim(x→1) x(x+1)(x+2)

  1. Apply the law twice. lim[(x(x+1))·(x+2)] = lim[x(x+1)] · lim(x+2) = [lim x · lim(x+1)] · lim(x+2).
  2. Evaluate each: 1 · 2 · 3 = 6.
Common mistake: expanding first (x³ + 3x² + 2x) and differentiating — wrong chapter! For limits, splitting is already the fast path.
Your turn: limx→2 x(x−1)(x+1) = ?

Answer: 6.

Apply the law twice: 2·1·3 = 6.

Before reading on: in x·sin(1/x) near 0, which factor breaks the product law — and what tool replaces it?

Example 4 — the trap: lim(x→0) x·sin(1/x)

  1. Check the hypothesis. lim(x→0) sin(1/x) does not exist. The product law is illegal here.
  2. Switch tools: squeeze. Since −1 ≤ sin(1/x) ≤ 1, multiply by |x|: −|x| ≤ x·sin(1/x) ≤ |x|.
  3. Outer limits match: both −|x| and |x| tend to 0, so the limit is 0.
Common mistake: writing “0 · DNE = 0” via the product law. The answer 0 is correct, but only the squeeze makes it legal. Examiners deduct for the illegal split.
Your turn: limx→0 x²·sin(1/x) = ?

Answer: 0.

The product law is illegal (sin(1/x) oscillates), so squeeze: −x² ≤ x²sin(1/x) ≤ x² → 0.

Memorization tips

  • The law goes one way: factors converging implies the product converges — but a convergent product does not imply convergent factors. x·sin(1/x) → 0 is the counterexample to memorize.
  • Bounded × vanishing = 0: the pattern behind every squeeze-theorem product. When you see oscillation times something → 0, think squeeze, not product law.
  • The add-and-subtract trick reappears: fg − LM = f(g−M) + M(f−L) is the same “add zero in disguise” move as the product rule proof. One trick, two theorems.
  • Boundedness is the hidden hypothesis: a convergent function is bounded near the point — that is what lets the proof control the |f|·|g−M| term.
  • Constant multiple is a special case: set g(x) = c. Fewer laws to memorize when you see the nesting.
  • Count your applications: n factors need n−1 uses of the law. Three factors, two splits.

Final challenge

Five mixed questions — basics, applications, and the traps, all in one. Score 5/5 and this law is yours.

← Back to the Calculus I formula sheet

How to learn a formula here

  1. Read each section in order — every section ends with a short quiz. Take it before moving on; the questions test exactly what you just read.
  2. Work the examples with the answers covered, then uncover one step at a time and compare.
  3. Finish with the final challenge — five mixed questions including the classic traps.
  4. Retake what you miss — every quiz reshuffles each attempt, and every answer explains itself.

Frequently asked questions

What is the product law for limits?

If lim(x→a) f(x) = L and lim(x→a) g(x) = M both exist and are finite, then lim(x→a) [f(x)·g(x)] = L·M: the limit of a product is the product of the limits.

Can a product's limit exist when a factor's limit doesn't?

Yes — lim(x→0) x·sin(1/x) = 0 even though lim(x→0) sin(1/x) DNE. The product law goes one way only; this case needs the squeeze theorem.

What is the add-and-subtract trick in the proof?

We write fg − LM = f·(g−M) + M·(f−L) (adding −fM + Mf = 0). Each group pairs one factor's error with the other's size, so both vanish in the limit.

Why does the proof need f to be bounded near a?

The error term |f|·|g−M| could blow up if |f| were unbounded. Convergent functions are bounded near the point, which keeps this term under control.

How do I handle three or more factors?

Apply the law repeatedly: lim(f·g·h) = lim((f·g)·h) = L·M·N, as long as every factor's limit exists.

More from the codex