Two Path Test: Definition, Examples, and Easy Step-by-Step Guide
The two-path test is a targeted technique in multivariable calculus for showing that a limit does not exist. Instead of trying to prove that a function approaches a single value from every possible direction, you compare its behavior along two specific paths that approach the same point. If those paths produce different limiting values, the overall limit cannot exist.
This test is especially useful for functions of two variables, where approaching a point like (0,0) can happen along infinitely many routes. A straight line, a parabola, or a spiral may each tell a different story. The two-path test gives you a fast way to catch contradictions before investing time in a more advanced proof.
What Is the Two-Path Test in Multivariable Calculus?
At its core, the two-path test is a logical shortcut based on the definition of a multivariable limit. For a limit to exist at a point, the function must approach the same real number no matter how the input approaches that point. If two valid paths give two different results, the function cannot have a single limiting value.
This idea is powerful because it turns an infinite-dimensional problem into a manageable comparison. You do not need to examine every possible curve or sequence. You only need to find one pair of paths that disagree. That single disagreement is enough to prove that the limit fails to exist.
The Formal Definition and Logic
Suppose a function f(x,y) is defined near a point (a,b), except possibly at (a,b) itself. Let C1 and C2 be two paths that both approach (a,b). If the limit of f along C1 is L1 and the limit along C2 is L2, and L1 is not equal to L2, then the two-path test proves that the overall limit does not exist.
The reasoning is a contrapositive argument. If the full limit existed and equaled L, then every path approaching (a,b) would have to approach L. Since the two paths approach different values, no single L can satisfy all paths. The limit therefore cannot exist.
What the Test Can and Cannot Prove
The two-path test is a one-way tool. Different limiting values along two paths prove nonexistence, but the same limiting value along two paths proves nothing. Two agreeing paths are only a hint, not a guarantee, because another path might still produce a different result.
Even if dozens of paths agree, a strange curved path or a sequence of points could disagree. To prove that a limit exists, you need a positive argument such as the squeeze theorem, polar coordinates with a bound independent of angle, algebraic inequalities, or an epsilon-delta proof. The two-path test is best used to falsify, not to verify.
How to Use the Two-Path Test Effectively
Start by confirming that the point in question is a valid limit point for the function. Then choose two paths that approach that point and lie inside the domain of the function, at least eventually. The goal is to simplify the expression enough to compute a one-variable limit along each path.
After computing both limits, compare the results. If they are different finite numbers, or if one path gives a finite number while the other grows without bound, the multivariable limit does not exist. If both paths give the same finite value, you must continue with a stronger method.
Choosing Paths That Reveal Different Limits
Common paths include the x-axis, the y-axis, lines of the form y = mx, parabolas such as y = x squared, and curves like y = kx squared. The best choice depends on the algebraic structure of the function. If the numerator and denominator contain terms of similar degree, a line y = mx often exposes dependence on the slope m.
For example, if the function contains x to the fourth power and y squared, a parabolic path like y = x squared may cancel terms differently than a straight line. If the resulting limit depends on m or on the shape of the path, then the original limit does not exist. If it does not depend on the path, the test remains inconclusive.
Interpreting Results and Avoiding Common Pitfalls
When two paths give different finite values, the conclusion is immediate: the limit does not exist. The same conclusion follows if one path approaches a finite number while another path approaches positive or negative infinity. However, if both paths give the same finite value, you cannot yet say the limit exists.
Common mistakes include using paths that do not actually approach the target point, forgetting to check whether the path lies in the function's domain, and assuming that two matching paths are enough for a proof. Another pitfall is algebraic cancellation that is invalid near the limit point. Always verify each simplification and remember that the two-path test is only a negative test.
Worked Examples of the Two-Path Test
Worked examples make the logic concrete. In each case, the strategy is to select paths that simplify the function enough to compute a one-variable limit. The goal is not to test every possible direction, but to find a pair of directions that disagree.
These examples also highlight why straight lines alone are sometimes insufficient. A function may behave consistently along every line through the origin and still fail to have a limit because a curved path reveals a different value. That is where the two-path test becomes especially valuable.
Example 1: xy/(x^2+y^2) at the Origin
Consider the function f(x,y) = xy divided by x squared plus y squared, and examine its limit as (x,y) approaches (0,0). Along the x-axis, set y = 0. Then f(x,0) = 0 divided by x squared, which equals 0 for every nonzero x, so the limit along this path is 0.
Now approach along the line y = x. Then f(x,x) = x squared divided by 2x squared, which simplifies to one half for every nonzero x. The limit along this path is one half. Since 0 and one half are different, the two-path test proves that the multivariable limit does not exist.
Example 2: x^2y/(x^4+y^2) at the Origin
Consider g(x,y) = x squared times y divided by x to the fourth power plus y squared. Along the x-axis, set y = 0. The numerator becomes 0 and the denominator becomes x to the fourth power, so the limit along this path is 0.
Now approach along the parabola y = x squared. Substituting gives a numerator of x to the fourth power and a denominator of x to the fourth power plus x to the fourth power, which is 2x to the fourth power. The ratio simplifies to one half, so the limit along the parabola is one half. Because the two paths disagree, the limit does not exist.
With enough practice, the two-path test becomes less about guessing and more about reading the algebra of a function. When you see terms that might cancel differently along a curve than along a line, try a strategic pair of paths and let the contradiction speak for itself. That habit will sharpen your intuition for limits, continuity, and the surprising geometry of multivariable functions.