Intuitive concept of a limit — table of values, graphical: Visual walkthrough

We are re-deriving the central idea of the parent topic: a limit is the destination the outputs aim at, not the value at the point.

Step 1 — Meet the machine and the broken gear

WHAT. A function is a machine: you feed it a number $x$ , it spits out a number we call $f (x)$ . Our machine is $f (x) = \frac{x ^{2} - 1}{x - 1} .$ Here $x$ is the input (the number we pour in), $x^{2} - 1$ is the top (numerator — it gets computed first), and $x - 1$ is the bottom (denominator — we divide by it).

WHY. Before asking "where does it head?", we must find where the machine jams. It jams whenever the bottom is $0$ , because dividing by zero is meaningless.

Set the bottom to zero: $x - 1 = 0 \Rightarrow x = 1$ . At exactly $x = 1$ the top is $1^{2} - 1 = 0$ too, so we get $\frac{0}{0}$ — the famous indeterminate form (see Indeterminate forms and algebraic simplification). Not "zero", not "infinity" — genuinely undecided.

PICTURE. The machine has one jammed gear, sitting exactly at $x = 1$ .

Figure — Intuitive concept of a limit — table of values, graphical

Step 2 — Sample near the jam and plot the dots

WHAT. Since we cannot stand on $x = 1$ , we stand just beside it and read the output. Pick $x = 0.5$ : top $= 0.25 - 1 = - 0.75$ , bottom $= 0.5 - 1 = - 0.5$ , so $f (0.5) = \frac{- 0.75}{- 0.5} = 1.5$ . Pick $x = 1.5$ : top $= 1.25$ , bottom $= 0.5$ , so $f (1.5) = 2.5$ .

WHY. We are gathering evidence. Each safe input gives one honest dot $(x, f (x))$ . Plotting many such dots reveals the shape the machine traces — the shape will tell us the destination.

PICTURE. Blue dots for real samples on the left and right of $x = 1$ ; the gap at $x = 1$ stays empty because that input is forbidden.

Step 3 — The algebra that reveals the hidden line

WHAT. Factor the top. The top $x^{2} - 1$ is a difference of two squares: $x^{2} - 1 = (x - 1) (x + 1)$ . Check: $(x - 1) (x + 1) = x^{2} + x - x - 1 = x^{2} - 1$ . ✓ So $f (x) = \frac{( x - 1 ) ( x + 1 )}{x - 1} .$ Here $(x - 1)$ is the shared factor living in both top and bottom, and $(x + 1)$ is the leftover after we remove it.

WHY. We want to cancel $(x - 1)$ top and bottom, turning the ugly fraction into something readable. Cancelling is legal only when $x - 1 \neq = 0$ — you may never cancel a zero. And here is the beautiful loophole: a limit sends $x \to 1$ but never lets $x = 1$ , so throughout the limit $x - 1 \neq = 0$ . The cancellation is allowed exactly because of the "near, not at" rule.

$f (x) = the leftover line (x + 1) for all x \neq = 1.$

PICTURE. The tangled fraction on the left; the clean straight line $y = x + 1$ on the right, with a single open circle (hollow dot) marking the one forbidden input $x = 1$ .

Step 4 — Where is the hole, and how high is it?

WHAT. Our machine behaves exactly like the line $y = x + 1$ , except the input $x = 1$ is banned. What height would that line have there? Plug $x = 1$ into the leftover: $x + 1 = 1 + 1 = 2$ . So the missing point sits at height $2$ .

WHY. We want to know the height the curve is aiming at. The line has no gap — it's a straight ruler pointing at $(1, 2)$ . Removing one point from a straight line does not bend it: the neighbours on both sides still line up toward $2$ . That target height is the candidate for our limit.

PICTURE. The line $y = x + 1$ with a hollow yellow circle at $(1, 2)$ — the exact spot the curve is missing but pointing at.

Step 5 — Squeeze from the left, squeeze from the right

WHAT. Now we creep in for real, from both sides, and watch the output heights.

$x \to 1^{-}$ (left)	$f (x)$	$x \to 1^{+}$ (right)	$f (x)$
$0.9$	$1.9$	$1.1$	$2.1$
$0.99$	$1.99$	$1.01$	$2.01$
$0.999$	$1.999$	$1.001$	$2.001$

The symbol $x \to 1^{-}$ means " $x$ slides up toward $1$ staying below it"; $x \to 1^{+}$ means " $x$ slides down toward $1$ staying above it" (these are the One-sided limits and limits at infinity).

WHY. The parent's rule: a two-sided limit exists only if both one-sided limits agree. Left heights climb $1.9 \to 1.99 \to 1.999 \to \dots$ toward $2$ . Right heights fall $2.1 \to 2.01 \to 2.001 \to \dots$ toward $2$ . The two marchers meet at the same wall.

PICTURE. Red arrow marching in from the left, green arrow marching in from the right, both squeezing onto the height $2$ at the hole.

Step 6 — Both sides agree, so the limit exists

WHAT. Left-hand limit $= 2$ . Right-hand limit $= 2$ . They match, so the full (two-sided) limit exists and equals that shared value: $lim_{x \to 1} \frac{x ^{2} - 1}{x - 1} = 2.$ Read the symbols: $lim$ = "the value approached", $x \to 1$ = "as input creeps to $1$ ", the fraction = "of this machine", $= 2$ = "is the number $2$ ".

WHY. This is the payoff of Steps 1–5 stitched together: the machine jams at $1$ (Step 1), but its neighbours (Step 2) trace a straight line (Step 3) pointing at height $2$ (Step 4), and both marchers confirm it (Step 5). Nothing was assumed — every claim came from a picture.

PICTURE. The final verdict: both arrows fused into one destination at $(1, 2)$ , with the equation stamped beside it.

Recall Why couldn't we just plug in

$x = 1$ ? Plugging in gives $\frac{0}{0}$ , undefined. Direct substitution is only a first guess; at a $0/0$ trap it fails, so we factor, cancel (legal since $x \neq = 1$ ), and read the destination instead.

Step 7 — The degenerate case: what if the sides disagreed?

WHAT. Our example was kind — both sides met. To know we earned the answer, we must see the case where the recipe refuses to give a limit. Take the jump function $g (x) = {12 x < 0 x \geq 0.$ From the left, $g \to 1$ . From the right, $g \to 2$ .

WHY. This proves "both sides must agree" is a real condition, not a formality. When the left marcher aims at height $1$ and the right marcher aims at height $2$ , there is no single destination — so $lim_{x \to 0} g (x)$ does not exist. Contrast this with Step 6, where agreement created the limit.

PICTURE. A cliff: left piece flat at height $1$ , right piece flat at height $2$ , the two arrows pointing at different walls.

The one-picture summary

Everything at once: the jammed fraction $\to$ the factor-and-cancel $\to$ the straight line with a hole $\to$ two arrows squeezing onto height $2$ $\to$ the boxed answer. Compare it to the failing jump so the contrast is burned in.

Recall Feynman retelling — explain the whole walkthrough to a 12-year-old

We had a math machine that jams if you feed it exactly $1$ — you'd be dividing by zero. So we sneak up on $1$ instead of standing on it. When we clean up the machine with a little algebra, it turns out to be the plain straight ruler " $x + 1$ " everywhere except that one banned spot. A straight ruler through the neighbours of $1$ points dead at the height $2$ . We walk toward $1$ from the left and from the right, and both walks aim at height $2$ — so even though there's a tiny missing tile at $(1, 2)$ , we know for sure that's where we were headed. The answer is $2$ . And we checked a mean example (a cliff) where the two walks aim at different heights: there, no single destination exists, so the limit doesn't exist. That's the whole trick — look where you're going, not where you're standing.

Where this leads next

The same "squeeze from both sides onto a value" idea powers The derivative as a limit of difference quotients — a difference quotient is a $\frac{0}{0}$ trap we tame exactly like this one.
When both-sides-agree is hard to prove directly, the Squeeze (Sandwich) Theorem traps the limit between two known curves.
To make "arbitrarily close" mathematically bulletproof, see the Formal epsilon-delta definition of a limit.

Flashcards

Why is

\frac{x ^{2} - 1}{x - 1}

undefined at

x = 1

The denominator

x - 1 = 0

there, giving the indeterminate form

\frac{0}{0}

What does

x^{2} - 1

factor into?

(x - 1) (x + 1)

(difference of squares).

Why may we cancel

(x - 1)

inside the limit?

Because

x \to 1

means

x \neq = 1

, so

x - 1 \neq = 0

and division by it is valid.

After cancelling, what simple function is left?

x + 1

, valid for all

x \neq = 1

What height does

y = x + 1

point at when

x = 1

2

Left-hand and right-hand limits of

\frac{x ^{2} - 1}{x - 1}

1

Both equal

2

, so the two-sided limit is

2

Why does the jump function

g

have no limit at

0

Left limit is

1

, right limit is

2

; they disagree, so no single destination exists.

Does the missing point (hole) at

(1, 2)

change the limit?

No — the limit tracks the approached height, not the value at the point.