Why Tables Can Mislead
A table can make a limit look easy to estimate: inspect some function values near a target input and look for a pattern. But a table is only a selection of inputs and outputs. A plausible pattern in a few entries is evidence, not a guarantee about every input close to the target.
Three errors are especially important to avoid. A reader may treat the function value at the target as the limit, combine values from opposite sides as though they describe one approach, or give distant values more weight than inputs very close to the target. Each error confuses what the table shows with what it can justify.
The Function Value Is Not the Limit
The limit of \(f(x)\) as \(x\) approaches \(a\) concerns the behavior of \(f(x)\) for inputs \(x\) close to \(a\), but not equal to \(a\). The value \(f(a)\), if it exists, is a separate piece of information. It may agree with the limit, differ from it, or be undefined.
That last sentence matters when reading tables. An entry at \(x=a\) reports \(f(a)\), not necessarily the limit. To estimate a limit, look at nearby inputs on the relevant side or sides.
Worked Example: Do Not Use One Entry at the Target
Suppose the function is defined by \(f(x)=x+1\) for \(x\ne2\), and \(f(2)=100\). Estimate the limit as \(x\) approaches 2 using the table.
At the target input, \(f(2)=100\). But for every other listed input, the rule is \(f(x)=x+1\). For example, \(f(1.99)=1.99+1=2.99\), \(f(2.01)=2.01+1=3.01\), and \(f(1.9)=1.9+1=2.9\). The nearby values on both sides are close to 3, so the table supports the estimate \(\lim_{x\to2}f(x)=3\), not 100.
Check: For any \(x\ne2\), \(f(x)=x+1\), so \(f(x)-3=x-2\). Thus, whenever \(x\ne2\) is within a distance \(\delta\) of 2, \(f(x)\) is within the same distance \(\delta\) of 3. The exceptional value \(f(2)=100\) does not affect this check.
Keep the Two Sides Separate
Inputs less than \(a\) approach \(a\) from the left; inputs greater than \(a\) approach from the right. A two-sided limit exists only when the function values approach the same number from both sides. Averaging left-side and right-side values can hide a disagreement instead of resolving it.
Here \(x\to a^-\) means that \(x\) approaches \(a\) through values less than \(a\), and \(x\to a^+\) means that \(x\) approaches through values greater than \(a\). A table with columns on both sides should be read in separate groups first.
Worked Example: Do Not Mix Opposite-Side Values
Let \(f(x)=2+x\) for \(x<0\), \(f(x)=-1+x\) for \(x>0\), and \(f(0)=7\). Use the table to decide whether a two-sided limit is supported.
On the left, \(f(-0.1)=2+(-0.1)=1.9\) and \(f(-0.01)=2+(-0.01)=1.99\). These values approach 2 as the negative inputs approach 0. On the right, \(f(0.01)=-1+0.01=-0.99\) and \(f(0.1)=-1+0.1=-0.9\). These values approach \(-1\) as the positive inputs approach 0.
The one-sided estimates disagree: the left side approaches 2 and the right side approaches \(-1\). Therefore, the table does not support a two-sided limit; in fact, the two-sided limit does not exist. The value \(f(0)=7\) is not the explanation for this failure: the disagreement comes from the different nearby behavior on the two sides.
Check: For negative \(x\), \(f(x)-2=x\), so the difference from 2 shrinks as \(x\) approaches 0 from the left. For positive \(x\), \(f(x)-(-1)=x\), so the difference from \(-1\) shrinks from the right. These rules confirm the two different one-sided estimates.
Give Nearby Inputs More Weight
A value can be near the target compared with the rest of the table and still be too far away to describe the behavior closer in. In general, a distant entry may show how the function behaves at that particular input, but it does not settle what happens arbitrarily close to the target. A convincing table should include inputs that get progressively closer from the side or sides being investigated.
For example, if the target is 0, inputs such as \(0.5\) and \(0.25\) are farther away than \(0.1\), \(0.01\), and \(0.001\). If the values change as the inputs get closer, that change is important evidence. Do not choose a limit estimate based mainly on the first, most distant entries.
Worked Example: Distant Entries Can Point the Wrong Way
Consider \(f(x)=4+\frac{1}{1+100x^2}\), with target input 0. Estimate the limit from positive inputs and compare the distant and nearby values.
For \(x=0.5\), \(100x^2=100(0.25)=25\), so \(f(0.5)=4+\frac{1}{26}=4.03846\ldots\), which rounds to 4.0385. For \(x=0.1\), \(100x^2=1\), so \(f(0.1)=4+\frac12=4.5000\). For \(x=0.01\), \(100x^2=0.01\), so \(f(0.01)=4+\frac{1}{1.01}=4.990099\ldots\), which rounds to 4.9901. For \(x=0.001\), \(100x^2=0.0001\), so \(f(0.001)=4+\frac{1}{1.0001}=4.999900\ldots\), which rounds to 4.9999.
The distant value at \(x=0.5\) is close to 4, but progressively nearer inputs produce values closer to 5. The nearby entries support an estimate of 5, not 4. To check that estimate directly, observe that \(f(x)-5=4+\frac{1}{1+100x^2}-5=-\frac{100x^2}{1+100x^2}\). As \(x\) gets close to 0, the numerator \(100x^2\) gets close to 0, while the denominator gets close to 1. This confirms that the function values get close to 5.
Check: Substituting \(x=0.001\) gives \(4+\frac{1}{1.0001}\), the table's value before rounding. The estimate of 5 matches the trend in the nearest entries, while the first entry alone would give a poor estimate.
Why a Few Entries Are Evidence, Not Proof
The definition of a limit requires nearby behavior, not merely a successful sample. More precisely, \(\lim_{x\to a}f(x)=L\) means that for every positive tolerance \(\varepsilon\), there is a positive distance \(\delta\) such that whenever \(0<|x-a|<\delta\), we have \(|f(x)-L|<\varepsilon\). The condition \(0<|x-a|\) excludes the target itself. The condition also concerns every input within the chosen distance, not just a few inputs printed in a table.
This definition explains why the common table-reading errors fail. The target entry is excluded. Left and right inputs must both be considered for a two-sided limit. And distant inputs may lie outside the small neighborhood where the limit behavior appears. A table can guide a good estimate, but it cannot by itself guarantee what every sufficiently close input will do.
Identify the input \(a\) and distinguish the entry \(f(a)\), if present, from nearby entries.
Read inputs below \(a\) and above \(a\) independently before making a two-sided estimate.
Compare the values for inputs closest to \(a\), and check whether they become more consistent as the inputs get closer.
If nearby values keep changing, or the two sides disagree, say that the table does not support a single two-sided estimate.
A Concrete Mistake and Its Correction
Suppose a table gives \(f(1.9)=2.9\), \(f(1.99)=2.99\), \(f(2)=100\), and \(f(2.01)=3.01\). A mistaken calculation might be: “The value closest to the target is \(f(2)=100\), so \(\lim_{x\to2}f(x)=100\).” That conclusion uses the function value at the target instead of nearby behavior.
The correction is to set aside \(f(2)\) and compare entries with \(x\ne2\). The values 2.9, 2.99, and 3.01 are close to 3, and the entries nearest to 2 are especially close to 3. Thus the table supports an estimate of 3. If the rule for the function is available, use it to check the estimate; if not, report it as an estimate supported by the table, not as a proof.
Using Tables Carefully
Before writing a limit estimate, ask what each group of entries actually says. Do not treat a single output as decisive. Do not combine opposite-side values until you have checked whether the sides agree. Do not let distant data outweigh a clear pattern among inputs closer to the target.
If entries are rounded, small differences may also be hidden by the displayed digits. A table should be read with its precision in mind: identical rounded values do not necessarily mean the underlying function values are exactly equal. The safest conclusion is the one justified by the direction, proximity, and displayed precision of the inputs and outputs.
Check Your Understanding
For each question, explain what the table supports and what it does not establish.
- A table lists \(f(3)=20\), while values at inputs just below and above 3 are close to 4. Which value is relevant to estimating \(\lim_{x\to3}f(x)\), and why?
- Values for \(x<a\) appear to approach 6, while values for \(x>a\) appear to approach 1. What can you conclude about a two-sided limit?
- Why should values at inputs nearest to the target generally receive more attention than values much farther away?
- What does the condition \(0<|x-a|\) exclude when defining a limit?
- Why can a finite table support a limit estimate without proving the limit?
