What Is a Limit in Calculus? Plain Meaning, Examples

Put x = 2 into (x² − 4) / (x − 2) and you get 0/0, which is not a number⁴. Yet the graph is a clean line with a single point missing, and every instinct says the missing height should be 4. A limit is the idea that turns that instinct into mathematics, and it is the idea under all of calculus²⁶.

What do limits mean in calculus?

A limit is the one real number that a function’s values approach as the input approaches a point, provided such a number exists². Read it as a statement about a journey, not a destination. You steer the input toward a target, and you watch where the output is heading.

In symbols you write the function, the point you are approaching, and the number L the outputs are heading toward³:

Notation: limit as x → a of f(x) = L³

Three words in that definition do the work². “Approach” means the outputs get closer and closer, from both sides. “One” means a single value, not a menu. “Provided such a number exists” means the whole thing can fail, which the page returns to below².

Limits are old. The limiting process, essential to understanding calculus, has been around for thousands of years². Early mathematicians used it to get better and better approximations of areas of circles². The formal definition used today did not appear until the late 19th century².

Limits are only concerned with what is going on around the point, and the point itself is excused³. Two functions that agree everywhere except at one input have the same limit there, which surprises nearly everyone the first time³.

A limit asks where the outputs are heading, never where the function happens to stand.

Worked example (our arithmetic): a hole in the graph

Take f(x) = (x² − 4) / (x − 2). Substituting x = 2 gives 0/0, so the function is undefined there, and that is not a problem, since the limit does not care what is happening at the point in question³. Tables of values are one way to see where the outputs head: choose values of x just below and just above the target². The arithmetic is ours, and the approach follows the textbook’s:

From below the outputs climb toward 4. From above they fall toward 4. When both sides head for a common value, that value is the limit². So the limit as x approaches 2 is 4, even though f(2) does not exist².

A table is a guess, and algebra turns the guess into a certainty. The numerator factors, and for every x other than 2 the common factor cancels⁴. Worked example (our arithmetic):

That is the plan for a whole family of limits. When a function has the 0/0 form, first make sure it cannot be evaluated directly with the limit laws, then find a simpler function that equals it for every x other than the target⁴. The diagram above is the picture of this exact function: the line y = x + 2, with an open circle at the missing point.

Why the limit ignores the point itself

Change the function at one point and nothing about the limit moves. Define a new function g with g(x) = (x² − 4) / (x − 2) for x other than 2, and give it the value 6 at x = 2. The function now has a value at 2, but the limit is NOT 6, because limits do not care what the function is actually doing at the point in question³. It stays 4.

So a limit and a function value are two different questions. The limit of a function can exist at a point, the function can be defined there, and the two numbers can still differ². Where they agree, the function is called continuous at that point, which in the notation of the course notes means the limit equals the value⁷.

For many everyday functions the two do agree, and a limit is easy: you can get a meaningful answer just by plugging in the limiting value⁷. A classic case is (x² + x) / (x + 1) at x = 3, where substitution gives 12/4, which is 3⁷. The harder limits are the ones where plugging in breaks, such as the 0/0 above.

When does a limit exist, and when does it not?

A limit exists when the outputs settle toward one real number from both sides. It does not exist in three ways, each with a picture.

• The two sides disagree. If the limit from the left and the limit from the right take different values, the limit of the function does not exist². The step function, often called the Heaviside function, does exactly this: from the right it heads for 1 and from the left for 0, so no single value is approached³.
• The values never settle. For a function that oscillates faster and faster near a point, the outputs do not approach any one single value². The lecture notes at MIT put a name on it: for something like this, we say the limit does not exist⁷. A table can even fool you here, because different input sets give different guesses³.
• The outputs grow without bound. For 1/x near zero, the values decrease without bound from the left and increase without bound from the right²⁷. Writing that the limit is infinity describes the behavior of the function, and it is not asserting that a limit exists².

The rule that ties the first case to the definition is two-sided: the limit exists at a point exactly when the limit from the left and the limit from the right have a common value². Failure is not a bug in the idea. Limits do not always exist, so do not get into the habit of assuming that they will³.

How to find a limit, step by step

• Try substitution. If the function is defined and well behaved at the point, plug the value in⁷.
• If you get 0/0, factor and cancel. Polynomials usually factor, and the shared factor cancels for every x other than the target⁴.
• If square roots are involved, use the conjugate. When the numerator or denominator contains a difference involving a square root, multiply the numerator and denominator by the conjugate of that expression⁴.
• Use tables and graphs only to understand, not to decide. They give insight, but they rely too much on guesswork², and using tables of values to guess a limit is simply not a good way to get the value of a limit³.

The limit laws are what make moves two and three legal. They are the individual properties of limits, such as the sum and product laws, and they let you break a limit into pieces⁴. Apply them one step at a time, because each new limit must exist for the law to apply⁴.

Worked example (our arithmetic): a full limit using substitution and then factoring. The function comes from the textbook⁴:

The limit is 3/7⁴.

The precise definition, in one worked case

The intuitive picture is enough for most of a first course. The formal definition makes “approach” exact. It is quite possibly one of the most challenging definitions you will encounter early in your study of calculus, but it is well worth the effort to reconcile it with the intuitive notion⁵. Understanding it is the key that opens the door to a better understanding of calculus⁵.

The definition is a game. A skeptic names how close the outputs must stay to L, called epsilon. You answer by naming how close the inputs must stay to the target, called delta. The limit is correct if you can win for every positive epsilon⁵. The textbook flags the phrase “there exists” as a signal for a scavenger hunt: you must go and find delta⁵.

Worked example (our arithmetic): the limit of 2x + 1 as x approaches 1, which is 3. The delta comes from the algebra the textbook uses for this limit, delta equal to epsilon over 2⁵:

Shrink epsilon and delta shrinks with it, and that is what “as close as you like” means. The history behind the machinery is short. Bernard Bolzano developed the basics of the epsilon-delta technique in 1817 to define continuous functions⁶. The modern notation, with the arrow written below the limit symbol, was invented in 1905⁶.

Where limits show up outside the classroom

Limits are the engine behind both halves of calculus. A tangent to a curve at a point is the limit of secant lines, which makes the slope a limit, and that slope is the derivative¹. Instantaneous velocity is the limit of average velocity over a small time interval¹. As the time interval shrinks, the average velocity becomes closer to the instantaneous velocity¹. The same idea explains why rate of change is one of the most critical concepts in calculus¹. The whole story is on the page about what a derivative is.

A limit can also describe a physical ceiling. In the textbook’s relativity example, an object’s mass grows without bound as its speed approaches the speed of light, and at one very high speed its mass becomes 7071 kg². Since no object can have an infinite mass, no object can travel at or more than the speed of light². That is a limit doing the work of a law of nature.

Common mistakes

• Treating the limit as the value at the point. They can differ, and the limit of the changed function above is still 4³.
• Reading 0/0 as “no limit”. It means the direct route failed and more algebra is needed⁴.
• Trusting a table. There is always the possibility that you chose the wrong values and guessed incorrectly³.
• Calling infinity a limit that exists. An infinite statement describes behavior and does not assert that a limit exists².
• Forgetting the sides. A single disagreement between left and right is enough to break the limit².

Where to go next

The derivative is the first big limit, the limit of slopes of secant lines, and the best next page is what a derivative is. The area under a curve is the other big limit, covered in what an integral is. The derivative rules cheat sheet shows what the limit machinery buys you once the rules are memorized. For the idea in motion, see how a normal morning hides rates and totals everywhere, and the rest of the core ideas section grows from here. The open texts behind this page are free to read: OpenStax, Lamar University and MIT OpenCourseWare.