Direct Substitution in Multivariable Limits

Why Direct Substitution Works

When a formula is built from familiar functions, its limit may be found simply by evaluating the formula at the point being approached. This procedure is called direct substitution. It is useful because it turns a limit calculation into an ordinary function evaluation—but it is valid only when the expression has the continuity and domain properties needed at that point.

For example, a polynomial in $x$ and $y$ can be evaluated at any point, and its values vary continuously as the inputs vary. A rational expression is also continuous wherever its denominator is nonzero. By contrast, if direct substitution makes a denominator zero, the expression is not defined there, and substitution alone does not determine the limit.

Central rule: If an expression is continuous at the point being approached, its limit there equals its value at that point. Before substituting, verify that the expression is defined there and that the continuity conditions for its components are satisfied.

Continuity and the Limit Statement

Let $D$ be the domain of a function $f:D\to\mathbb{R}$, and let $\mathbf{a}$ be a point that can be approached by points of $D$ other than $\mathbf{a}$. The statement $\lim_{\mathbf{x}\to\mathbf{a}}f(\mathbf{x})=L$ means that function values approach $L$ as domain points $\mathbf{x}$ approach $\mathbf{a}$, with $\mathbf{x}\ne\mathbf{a}$.

For direct substitution, the key condition is continuity at $\mathbf{a}$: the function must be defined at $\mathbf{a}$, and its values near $\mathbf{a}$, within its domain, must approach $f(\mathbf{a})$. In that case, the limit is exactly $f(\mathbf{a})$. This also applies when the domain has a boundary at $\mathbf{a}$, provided the limit is taken through points of the domain and the function is continuous there relative to its domain.

Direct substitution theorem: If $f$ is continuous at $\mathbf{a}$, then $$ \lim_{\mathbf{x}\to\mathbf{a}} f(\mathbf{x})=f(\mathbf{a}). $$ For a formula composed of elementary functions, this gives a practical test: substitute the coordinates of $\mathbf{a}$, provided every operation is defined and continuous there on the relevant domain.

Which Expressions Allow Substitution?

Polynomials in several variables are continuous at every point of $\mathbb{R}^n$. Sums, differences, and products of continuous expressions remain continuous wherever the expressions are defined. A quotient is continuous at points where its denominator is nonzero. These are the standard continuity facts underlying the direct-substitution rule.

The elementary functions used in formulas also have domain conditions. The exponential function is continuous for every real input. Sine and cosine are continuous for every real input. The real square-root function is continuous on $[0,\infty)$, so a formula containing a square root requires a nonnegative radicand wherever that formula is evaluated. If a square root is composed with another expression, that inner expression must remain in the square-root domain for the nearby domain points under consideration.

More generally, for a composition $h(g(\mathbf{x}))$, the values $g(\mathbf{x})$ must lie in the domain of $h$ for the nearby points being considered. It is not enough that $h$ is continuous at $g(\mathbf{a})$ if the composition is undefined at nearby inputs. When this domain requirement holds and the component functions are continuous at the relevant inputs, the composition is continuous and direct substitution applies.

Worked Example: A Polynomial Limit

Evaluate $\lim_{(x,y)\to(1,-2)}(2x^2y-3xy+4y^2)$. The formula is a polynomial, so it is continuous at $(1,-2)$. Substitute $x=1$ and $y=-2$:

\begin{aligned} 2(1)^2(-2)-3(1)(-2)+4(-2)^2 &=2(1)(-2)+6+4(4)\\ &=-4+6+16\\ &=18. \end{aligned}

Therefore the limit is $18$. Checking the evaluation term by term gives $-4$, $6$, and $16$, whose sum is $18$; this is the polynomial's value at the target point.

Worked Example: A Rational Limit with a Nonzero Denominator

Evaluate $\lim_{(x,y)\to(2,1)}\frac{x^2+xy+2y^2}{x+3y}$. The denominator at the target point is $2+3(1)=5$, which is nonzero. Thus the rational expression is defined and continuous there, so substitution is valid.

\begin{aligned} \frac{x^2+xy+2y^2}{x+3y} \bigg|_{(x,y)=(2,1)} &=\frac{(2)^2+(2)(1)+2(1)^2}{2+3(1)}\\ &=\frac{4+2+2}{2+3}\\ &=\frac{8}{5}. \end{aligned}

The limit is $\frac{8}{5}$. A direct check of the substituted numerator and denominator gives $8$ and $5$, respectively, so their quotient is indeed $\frac{8}{5}$.

Radicals and Domain Boundaries

A square-root expression needs a little extra care. The radicand must be nonnegative at each point where the real-valued formula is used. If the radicand is positive at the target point, it remains positive for all sufficiently nearby points, by continuity of the radicand. If the radicand equals zero at the target point, the target may be on the boundary of the expression's domain. In that case, the limit is taken through the domain, and continuity of the square root on $[0,\infty)$ still permits substitution.

Worked Example: A Radical at the Boundary of Its Domain

Consider $\lim_{(x,y)\to(0,0)}\sqrt{x+y}$, with the function defined on $D=\{(x,y):x+y\ge0\}$. The target $(0,0)$ belongs to $D$, and the radicand there is $0$. The limit is taken through points in $D$, so every radicand used in evaluating the function is nonnegative.

\begin{aligned} \lim_{\substack{(x,y)\to(0,0)\\(x,y)\in D}}\sqrt{x+y} &=\sqrt{0+0}\\ &=0. \end{aligned}

This substitution is valid because $u\mapsto\sqrt{u}$ is continuous on its domain $[0,\infty)$, including at $u=0$, and $x+y\to0$ through nonnegative values. The resulting function values approach $\sqrt{0}=0$.

Combining Elementary Functions

Many formulas combine several types of functions. The same procedure applies: check each operation's domain, then substitute. For a square root, check the radicand; for a quotient, check the denominator; for an exponential or trigonometric function, the real input is allowed throughout its domain. If all these checks pass and the component functions are continuous at the relevant inputs, the whole expression is continuous at the target.

Worked Example: Radical, Exponential, and Trigonometric Factors

Evaluate $\lim_{(x,y)\to(1,1)}\left(\sqrt{4+x-y}+e^{x-y}\sin\left(\frac{\pi y}{2}\right)\right)$. At $(1,1)$, the radicand is $4+1-1=4>0$, so the square root is defined and continuous near the point. The exponential, sine, and their inputs are also defined and continuous there.

\begin{aligned} \sqrt{4+1-1}+e^{1-1}\sin\left(\frac{\pi(1)}{2}\right) &=\sqrt{4}+e^0\sin\left(\frac{\pi}{2}\right)\\ &=2+(1)(1)\\ &=3. \end{aligned}

Thus the limit is $3$. Substitution into the original expression gives the same value: the radical term is $2$, and the product term is $1$, for a total of $3$.

Why the Rule Works

The direct-substitution rule is a consequence of the definition of continuity. If $f$ is continuous at $\mathbf{a}$, then for every $\varepsilon>0$, there is a $\delta>0$ such that domain points satisfying $\|\mathbf{x}-\mathbf{a}\|<\delta$ also satisfy $|f(\mathbf{x})-f(\mathbf{a})|<\varepsilon$. This is exactly the condition that $f(\mathbf{x})$ approach $f(\mathbf{a})$ as $\mathbf{x}$ approaches $\mathbf{a}$.

For a formula made from continuous components, the continuity laws ensure that the output remains close to its value at the target when the inputs are close to the target. The domain checks matter because those laws only apply when the relevant operations are defined. For example, a quotient cannot be treated as continuous at a point where its denominator is zero.

Check the target point.
Confirm that it belongs to the function's domain, or that the limit is explicitly being taken through a domain containing nearby points.

Check every operation.
Verify that denominators are nonzero and real square-root radicands are nonnegative at the point.

Use continuity.
When the component functions are continuous at their relevant inputs and the composition is defined nearby, evaluate the formula at the target.

When Substitution Does Not Decide the Limit

A common mistake is to substitute into a quotient, obtain an undefined expression, and treat that output as the limit. Consider $\frac{x^2-y^2}{x-y}$ at $(0,0)$. Direct substitution gives $\frac{0}{0}$, which is not a number. It does not mean that the limit equals zero, and it does not by itself mean that the limit fails to exist. It means only that the quotient formula is undefined at the target and the direct-substitution theorem cannot be applied there.

Common pitfall: The condition for direct substitution is not merely “the formula looks familiar.” The expression must be defined and continuous at the target. A zero denominator or an invalid radical input prevents this method from establishing the limit.

The same caution applies to nested expressions. If one substitutes into an inner function and obtains an input outside the outer function's domain, then the composed expression may be undefined. A continuity claim about the outer function at one value cannot justify evaluating a composition at nearby inputs that do not belong to its domain. Always check that the relevant function values exist before using continuity.

Summary

Direct substitution evaluates a multivariable limit at a point where the expression is continuous. Polynomial formulas allow it everywhere; rational formulas allow it where the denominator is nonzero; radicals require valid radicands; and exponential and trigonometric functions accept real inputs. For compositions, the intermediate values must stay in the relevant domains. When substitution produces an undefined expression, it gives no conclusion about the limit.

Check Your Understanding

For each question, decide whether direct substitution is justified before calculating.

What domain condition must be checked before substituting into a rational expression?
Evaluate $\lim_{(x,y)\to(1,2)}(x^2+3xy-y)$, and state why substitution applies.
For $\sqrt{2x-y}$ at $(1,2)$, what is the radicand at the target, and how does the domain affect the limit?
What does the result $\frac{0}{0}$ after substitution tell you—and what does it not tell you?
Why must the intermediate values of a composition lie in the outer function's domain near the target?