AP Calculus AB/BC · Unit 1 · Topic 1.9
Connecting Multiple Representations of Limits
Translate the same nearby behavior among formulas, graphs, tables, and verbal descriptions.
1. Topic Focus
Build the language of limits, connect numerical, graphical, and algebraic representations, and use continuity theorems with verified hypotheses.
This topic: Translate the same nearby behavior among formulas, graphs, tables, and verbal descriptions.
2. Key Relationship
Read every symbol with its domain, direction, units, and hypotheses before applying the relationship.
3. Visual Connection
4. Worked Example
The formula, a line with a hole, and a table all show a limit of 2 at x=1.
Write the governing relationship first, carry out the algebra cleanly, and finish with a sentence that answers the mathematical question.
5. Concept Development
One limit, four representations
The statement \(\lim\limits_{x\to a}f(x)=L\) can be communicated analytically, graphically, numerically, or verbally. A correct translation preserves the target input \(a\), the direction of approach, the approached output \(L\), and the distinction between nearby behavior and \(f(a)\).
| Representation | What carries the limit information | Typical strength | Typical limitation |
|---|---|---|---|
| Analytical | A formula, identity, inequality, or known component limits | Can justify an exact value | Domain restrictions may be hidden by simplification |
| Graphical | The heights approached by branches near \(x=a\) | Shows sides, holes, jumps, oscillation, and unbounded behavior | Window and resolution can conceal fine behavior |
| Numerical | Output trends for inputs approaching \(a\) from each side | Supports a precise estimate and detects side differences | Finite samples cannot prove all nearby behavior |
| Verbal | A description of how outputs behave as inputs approach | Makes direction and meaning explicit | Imprecise wording can confuse a limit with a function value |
Facts that must survive every translation
| Mathematical fact | Graph | Table | Words |
|---|---|---|---|
| \(x\to a^-\) | Trace the branch with \(x<a\) | Use inputs below \(a\) | Approach \(a\) from the left |
| \(x\to a^+\) | Trace the branch with \(x>a\) | Use inputs above \(a\) | Approach \(a\) from the right |
| \(f(x)\to L\) | Branch heights move toward \(L\) | Outputs settle near \(L\) | Nearby outputs approach \(L\) |
| \(f(a)\) | Read a filled point at \(x=a\) | Use a separate row at exactly \(a\) | State the assigned point value separately |
A filled point can move without changing a nearby formula, branch, or table trend. Therefore changing \(f(a)\) alone does not change \(\lim\limits_{x\to a}f(x)\).
Translate a finite two-sided limit
The following statements communicate the same nearby behavior:
- Symbolic: \(\lim\limits_{x\to a^-}f(x)=L=\lim\limits_{x\to a^+}f(x)\), so \(\lim\limits_{x\to a}f(x)=L\).
- Graphical: both branches approach the point height \(L\) on the vertical line \(x=a\), whether that point is open or filled.
- Numerical: outputs associated with increasingly close inputs below and above \(a\) both trend toward \(L\).
- Verbal: as the input approaches \(a\) from either side, the output approaches \(L\).
The graph need not cross or touch \((a,L)\), and table entries need not ever equal \(L\).
Translate disagreement and unbounded behavior
Representations must preserve one-sided information when a two-sided finite limit fails.
| Nearby behavior | Symbolic conclusion | Representation clues |
|---|---|---|
| Left approaches \(P\), right approaches \(Q\), and \(P\ne Q\) | Two-sided limit DNE | Graph has different branch heights; table has different side trends |
| Both sides increase without bound | \(\lim\limits_{x\to a}f(x)=+\infty\) | Branches rise beyond the window; numerical magnitudes grow positively |
| One side decreases without bound and the other increases without bound | Two-sided limit DNE | Opposite branch directions and opposite-sign numerical growth |
| Outputs continue oscillating through a fixed range | Limit DNE | Zoomed graphs keep oscillating; carefully chosen tables show multiple output clusters |
Infinity describes unbounded behavior, not a real function value. Always retain the approach direction when the sides differ.
Equivalent formulas and removable points
If two formulas agree for all sufficiently close \(x\ne a\), they describe the same limit at \(a\). For example,
The simplified formula predicts a line approaching height 4. The original graph is that line with the input \(x=2\) excluded unless another value is assigned. A table should omit 2 when estimating the limit but sample values on both sides of it.
Build a useful numerical representation
- Choose paired inputs below and above the target.
- Reduce their distances from the target, such as \(0.1,0.01,0.001\).
- Keep enough digits to distinguish a real trend from rounding noise.
- Compare the two sides independently before giving a two-sided conclusion.
- Do not include \(x=a\) as evidence for the limit; record \(f(a)\) separately if requested.
Use representations to check one another
When two representations appear to disagree, do not choose the more attractive answer. Investigate the source of the discrepancy:
- Formula: Did simplification silently remove a domain restriction?
- Graph: Is the window too wide, too narrow, or vertically clipped?
- Table: Are inputs approaching from both sides, and are they accidentally following a special sequence?
- Technology: Did rounding, degree mode, or finite precision create misleading values?
- Point value: Is \(f(a)\) being confused with the limit?
Analytical reasoning can justify an exact conclusion, while graph and table evidence can reveal what the algebra should explain.
A translation routine
- Record the target input and direction.
- Determine the left-side trend and write its one-sided notation.
- Determine the right-side trend and write its one-sided notation.
- Compare the sides to state the two-sided conclusion.
- Record \(f(a)\) separately.
- State whether the conclusion is exact, estimated, unbounded, or DNE.
- Translate into the requested representation without changing any of those facts.
6. Detailed Worked Example and Error Check
Example 1: A removable point in four representations. Define
Analytical: factoring gives \(f(x)=x+2\) for \(x\ne3\), so the exact limit is 5.
| \(x<3\) | \(f(x)\) | \(x>3\) | \(f(x)\) |
|---|---|---|---|
| 2.9 | 4.9 | 3.1 | 5.1 |
| 2.99 | 4.99 | 3.01 | 5.01 |
| 2.999 | 4.999 | 3.001 | 5.001 |
Graphical: the graph is the line \(y=x+2\) with an open point at \((3,5)\). Verbal: as \(x\) approaches 3 from either side, the outputs approach 5, although \(f(3)\) is undefined.
Example 2: Translate a jump. Let
The left formula approaches 3 and the right formula approaches 5:
A matching graph has an open left endpoint at \((1,3)\) and a right branch beginning at height 5. A matching table has left outputs near 3 and right outputs near 5. The filled value \(p(1)=5\) does not repair the jump.
Example 3: Combine information from different representations. A graph shows \(\lim\limits_{x\to4}u(x)=2\), while a two-sided table shows \(\lim\limits_{x\to4}v(x)=-1\). Then the limit laws give
The component limits can come from different representations, but their target inputs and approach directions must match before they are combined.
Example 4: Translate opposite unbounded sides. For \(r(x)=1/(x-2)\), sign analysis gives
The graph has a vertical asymptote at \(x=2\), descending on the left and rising on the right. A table shows increasingly large negative values from below and positive values from above. The two-sided limit is DNE because the sides do not share one behavior.
Example 5: A table can follow a misleading sequence. Consider
At the special inputs \(x=1+1/n\), the table always reports \(s(x)=\sin(n\pi)=0\). That table alone appears to suggest a limit of 0. Other inputs make the sine values approach 1 or \(-1\), and the graph keeps oscillating at every magnification. Therefore \(\lim\limits_{x\to1}s(x)\) is DNE. A reliable numerical investigation varies the input pattern rather than trusting one special sequence.
7. AP Reasoning Routine
Read one-sided behavior first, choose a matching limit procedure, and justify conclusions with definitions or theorem conditions.
- Identify the representation and requested quantity.
- State the rule or theorem and verify its conditions.
- Keep exact values until the final requested approximation.
- Interpret sign, units, interval, and context.
Connect the requested representations and justify each conclusion.
(a) For \(f(x)=\frac{x^2-25}{x-5}\), describe the equivalent nearby formula, the graph near \(x=5\), a suitable two-sided table trend, and the verbal meaning of the limit.
(b) A table has left-side outputs approaching \(-2\) and right-side outputs approaching 4 as \(x\to3\). Write both one-sided limits and the two-sided conclusion, then describe the matching graph behavior.
(c) Translate “outputs increase without bound as \(x\) approaches \(-1\) from the right” into notation and a graphical description.
(d) A graph has both branches approaching the open point \((2,6)\) and a filled point at \((2,-3)\). State the two-sided limit and function value separately.
(e) A graph gives \(\lim\limits_{x\to0}a(x)=3\), and a table gives \(\lim\limits_{x\to0}b(x)=2\). Find \(\lim\limits_{x\to0}\frac{a(x)+b(x)}{b(x)}\) and name the required condition.
(f) Describe a piecewise formula and table consistent with a graph whose left branch approaches 1 and right branch approaches 1 at \(x=0\), while the filled point is at height 5.
(g) A calculator graph of \((x^2-4)/(x-2)\) appears to be the complete line \(y=x+2\). Explain the missing graphical detail and how algebra reveals it.
(h) Explain why a table using only \(x=1+1/n\) cannot establish the limit of \(\sin(\pi/(x-1))\) as \(x\to1\).
Check the solution
In part (a), factoring gives \(x+5\) for \(x\ne5\), so the graph is the line \(y=x+5\) with a hole at \((5,10)\); paired table outputs approach 10 from both sides, meaning that nearby outputs approach 10 as \(x\to5\). In part (b), \(\lim\limits_{x\to3^-}f(x)=-2\) and \(\lim\limits_{x\to3^+}f(x)=4\), so the two-sided limit is DNE; the graph branches approach different heights. Part (c) is \(\lim\limits_{x\to-1^+}f(x)=+\infty\); the right branch rises without bound beside the vertical line \(x=-1\). In part (d), the limit is 6 and \(f(2)=-3\). In part (e), the quotient law applies because the denominator limit is \(2\ne0\), giving \((3+2)/2=5/2\). For part (f), one choice is \(f(x)=x+1\) for \(x\ne0\) and \(f(0)=5\); paired outputs near 0 approach 1. In part (g), the original formula excludes \(x=2\), so the line must have a hole at \((2,4)\); cancellation is valid only for \(x\ne2\). In part (h), those inputs sample only one special sequence and always produce 0, while other sequences produce different output clusters; finite or specially patterned samples cannot prove one nearby trend.