The Limit Laws (Properties of Limits)

When you first meet limits, the temptation is to treat every new function as a fresh problem and start plugging in numbers. That is slow, and it hides the pattern. The limit laws, also called the properties of limits, are the algebraic rules that let you break a complicated limit into smaller pieces you already know. They are not shortcuts around the definition; they are the definition, unpacked into operations you can perform without re-proving anything. In Stewart's 'Calculus', Section 2.3, these are stated as the Limit Laws, and they are the backbone of every limit you will compute by hand from here on. If you cannot use these, you cannot do calculus. If you can, most limit problems become a matter of checking a few conditions and then substituting.

Below is the complete set of limit rules you are expected to know. Each law assumes that the individual limits exist as finite real numbers. That condition matters more than the law itself, because the laws are only true when the pieces behave. The table shows the formula, the condition, and a plain-language reading of what it allows.

g(x)

Law	Formula	Condition
Sum	limx→a [f(x) + g(x)] = limx→a f(x) + limx→a	Both individual limits exist and are finite.
Difference	limx→a [f(x) − g(x)] = limx→a f(x) − limx→a g(x)	Both individual limits exist and are finite.
Constant Multiple	limx→a [c · f(x)] = c · limx→a f(x)	c is any real number; the limit of f exists.
Product	limx→a [f(x) · g(x)] = limx→a f(x) · limx→a g(x)	Both individual limits exist and are finite.
Quotient	limx→a [f(x) / g(x)] = limx→a f(x) / limx→a g(x)	Both individual limits exist, and limx→a g(x) ≠ 0.
Power	limx→a [f(x)]n = [limx→a f(x)]n	n is a positive integer. If n is any real number, the limit of f must be positive.
Root	limx→a ⁿ√f(x) = ⁿ√[limx→a f(x)]	n is a positive integer; if n is even, the limit of f must be positive.

The conditions are not decoration. The quotient law fails the moment the denominator heads to zero, even if the numerator also heads to zero. That 0/0 situation is an indeterminate form, and no amount of law-application will rescue it directly. You have to change the algebra first. Similarly, the power law with a fractional exponent requires the base limit to be positive; taking the square root of a negative number is not a real limit. The same goes for the root law with an even index. The sum and product laws fail if either piece oscillates without settling, as with sin(1/x) near zero. In that case, the individual limit does not exist as a real number, so the law is void. The common error is to apply the law mechanically and then wonder why the answer is wrong. Check the conditions before you substitute.

Take lim_x→2 (3x² + 5x − 1). Direct substitution gives 3(4) + 10 − 1 = 21, but the point of the exercise is to see the laws in action. By the sum law, split it into three limits: lim_x→2 3x², lim_x→2 5x, and lim_x→2 (−1). The constant multiple law pulls the 3 and 5 out front. The power law turns x² into [lim_x→2 x]² = 2² = 4. The constant law gives −1. Putting it together: 3·4 + 5·2 − 1 = 12 + 10 − 1 = 21. The answer matches direct substitution because the function is a polynomial, and polynomials are continuous everywhere. That is not luck; it is the reason the laws exist. They turn a limit problem into an arithmetic problem.

Why Direct Substitution Works Here and Not Everywhere

Direct substitution works for polynomials and rational functions at points where the denominator is not zero. The limit laws are what justify that shortcut. When substitution gives 0/0, the laws still apply, but not yet. You need to factor, cancel, or rationalize first. That is the difference between a limit that exists and one that does not. The laws do not tell you the value when the denominator vanishes; they tell you to stop and change the expression.

Two limits appear so often that memorising them saves time and prevents mistakes. The first is lim_x→0 sin(x)/x = 1. This is not a law; it is a proven result, and it is the foundation for the derivative of sin(x). The second is lim_x→0 (1 − cos(x))/x = 0. Both are proved using the squeeze theorem, which is a separate technique for limits that oscillate or are bounded between two others. If you see a trigonometric limit that gives 0/0, these two are your first suspects. They are not limit rules in the algebraic sense, but they are limit formulas you will use in every subsequent chapter. Learn them as facts, not as something to re-derive.

The constant multiple law is the one students forget because it feels too obvious. lim_x→a [c·f(x)] = c·lim_x→a f(x). It is a special case of the product law where g(x) is a constant function. It matters because it lets you pull out coefficients before you do any other work. In a limit like lim_x→3 7x², you pull the 7 out, then apply the power law to x². Without this rule, you would have to multiply first, which is slower and more error-prone. It also works in reverse: if you have a limit that equals a constant, you can factor that constant out of a larger expression. This is the workhorse of simplification, and it is the first law to apply when you see a coefficient.

The root law is the power law with a fractional exponent, but it carries an extra condition. lim_x→a ⁿ√f(x) = ⁿ√[lim_x→a f(x)] only makes sense when the limit inside is non-negative for even n. That is not a formality. If you take the fourth root of a limit that is −2, you are not doing real arithmetic. The limit of the root does not exist as a real number. For odd n, negative values are fine. The rule you should remember: if the index is even, check the sign of the inner limit before you apply the law. This is the kind of condition that separates a correct solution from a lucky one.

Stewart's note in Section 2.3 extends the power law beyond positive integers. If lim_x→a f(x) is positive, then n can be any real number, not just a whole number. That covers fractional and irrational exponents. The catch is the positivity condition. If the base limit is zero or negative, the law is not automatically valid. For example, lim_x→0 x^1/3 works because the cube root of zero is zero and the function is defined. But lim_x→0 x^1/2 does not exist as a real limit because the function is not defined for negative x, so the left-hand limit does not exist. The condition is not pedantic; it is the difference between a real answer and a non-existent one.

The quotient law is the one that causes the most trouble because the condition is strict. lim_x→a [f(x)/g(x)] = [lim_x→a f(x)] / [lim_x→a g(x)] requires that the denominator limit is not zero. If it is zero, you have an indeterminate form of type 0/0, and the law is off the table. What do you do then? You factor, cancel, or rationalize. That is a different technique, covered in the guide on calculating limits manually. The failure case is when students try to apply the quotient law to a 0/0 limit and get a wrong answer like 0 or 1. The correct move is to stop and simplify. The quotient law is for when the denominator settles on a non-zero number, not for when it vanishes.

The epsilon-delta definition is the formal foundation of limits. It says that for every ε greater than zero, there exists a δ such that if 0 < |x − a| < δ, then |f(x) − L| < ε. That definition is what proves the limit laws, but it is not a tool for evaluating limits. You do not use epsilon-delta to find lim_x→2 (3x² + 5x − 1); you use the laws. Epsilon-delta is for proving the laws are true, and for checking whether a limit exists in pathological cases where the function does not behave. If you are a student, you should know what it means, but you should not reach for it when solving problems. That is the gap between theory and technique, and the limit laws are the bridge.

When the quotient law fails because of 0/0, the manual techniques are factoring, rationalizing, and then using the squeeze theorem for oscillating functions. Factoring works when the numerator and denominator share a common factor that cancels. Rationalizing works when a radical is involved, by multiplying by the conjugate. The squeeze theorem is for functions trapped between two others that have the same limit. These are not limit laws; they are strategies for rewriting the expression so that the laws can apply. The order matters: first try substitution, then check if a law applies, then rewrite. If none of that works, the limit may not exist, or it may be an infinite limit.

The special limits like lim_x→0 sin(x)/x = 1 are proved using the squeeze theorem, not the limit laws. The laws assume the pieces have limits; the squeeze theorem establishes the limit when the function is bounded. This is a different tool for a different situation. You use the squeeze theorem when you cannot algebraically simplify your way to a limit, because the function oscillates or is trapped. The classic example is x²·sin(1/x) as x→0, which squeezes between −x² and x², both of which go to zero. Knowing when to switch from laws to squeeze is a skill, and it comes from recognizing the form of the problem, not from memorizing a formula.

A two-sided limit exists only if the left-hand and right-hand limits both exist and are equal. This is not a limit law; it is the definition of the limit. The laws assume the two-sided limit exists, so if you are unsure, check the one-sided limits first. For piecewise functions, this is the difference between a correct answer and a wrong one. At a boundary point, the function may approach different values from each side. The limit laws do not help you there because they require a single limit. You have to evaluate each side separately, and if they disagree, the limit does not exist. That is the failure case for the laws, and it is the one students miss most often.

The limit laws also apply when x approaches infinity, provided the individual limits exist as finite numbers. For example, lim_x→∞ (3/x + 2) uses the sum law and the constant law. But when you have a rational function like (2x² + 1)/(x² − 3), you cannot apply the quotient law directly because the limits are infinite. Instead, you divide every term by the highest power of x. That is a separate technique, and it is covered in the guide on limits at infinity. The laws still work, but only after you rewrite the expression. The failure case is trying to use the quotient law on an ∞/∞ form, which is indeterminate. Rewrite first, then apply the laws.

The old guide tried to cover epsilon-delta, the squeeze theorem, infinity, and one-sided limits all in one breath. That is too much, and it buries the limit laws, which are the actual subject. It also implied that the epsilon-delta definition is how students evaluate limits, which is false. Most students use the laws and algebraic techniques. The guide also treated ∞ as if it were a number, which contradicts the definition of a limit as a real number. A limit that grows without bound does not exist as a real number; it is a specific type of non-existence. The fix is to separate the laws from the formal definition and from the special cases.

The most common mistake is applying the quotient law when the denominator limit is zero. The second is forgetting the positivity condition on the power and root laws. The third is assuming the sum law works when one of the limits does not exist. Each of these errors comes from treating the laws as unconditional, when they are conditional on the pieces existing. A fourth mistake is confusing a limit that is infinite with a limit that does not exist. They are not the same. An infinite limit means the function grows without bound; a non-existent limit means it does not settle on any value, which includes oscillation. Write your answer as DNE for the latter, and as ∞ or −∞ only when the function actually grows without bound.

Most guides list the laws and stop, but the practical detail is that the quotient law's zero-denominator condition is the single most violated rule in first-year calculus, and the failure mode is almost always a student who substitutes 0/0 and writes 1 instead of stopping to factor.

What are the limit laws in calculus?

The limit laws are a set of algebraic rules, also called the properties of limits, that let you break a limit of a sum, product, quotient, or power into the limit of its parts, provided each part has a finite limit.

Can I use the quotient law if the denominator limit is zero?

No. The quotient law requires the denominator limit to be non-zero. If it is zero, you have an indeterminate form and must factor, cancel, or rationalize before applying any law.

What is the difference between a limit that is infinite and one that does not exist?

An infinite limit means the function grows without bound, written as ∞ or −∞. A limit that does not exist (DNE) means the function does not approach any single value, including oscillation or differing one-sided limits. ∞ is a specific type of DNE, not a number.

When can I apply the power law to a real exponent?

You can apply the power law with any real exponent n only if the limit of the base f(x) is positive. If the base limit is negative or zero, the law may fail, especially for fractional exponents.

Do the limit laws work for one-sided limits?

Yes, the laws apply identically to left-hand and right-hand limits, but only if the relevant individual one-sided limits exist. For the two-sided limit to exist, both one-sided limits must be equal.

Why is lim_x→0 sin(x)/x = 1 a special limit and not a law?

It is a proven result, not an algebraic rule. It is derived using the squeeze theorem and is used as a building block for evaluating other trigonometric limits. It is a limit formula you memorise, not a property of limits.

What is the most common mistake with the limit laws?

The most common mistake is applying the quotient law when the denominator limit is zero, leading to a false 0/0 answer. Always check the conditions of each law before substituting values.