Suppose that f ( x , y ) satisfies the following conditions:
Must the function f ( x , y ) be continuous?
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Your solution complicates the intention of the question. It provides a counter-example, but doesn't explain what's really going on.
The key point of the question is to realize that for the function to be continuous (say at the origin), we require the limit to exist along every possible path to the origin. However, we're only given 2 directions (x-axis, y-axis), and hence have very limited control over what happens anywhere else. This suggests that we could let the limit along the path y = K x , or the path y = K x 2 , be varying and hence the limit doesn't exist so the function isn't continuous.
For the path of y = K x , if we naively set x y , we do get a varying value of K , but end up with the x-axis being undefined. To fix that, we use a counter example of f ( x , y ) = x 2 + y 2 x y and f ( 0 , 0 ) = 0 . In this case, setting y = K x gives us 1 + K 2 K , which is clearly not equal for all values of K .
See if you can come up with a (simple) counter example for the path y = K x 2 . Even though this might seem to locally look like y = 0 at the origin, it's still possible to create an issue.
Log in to reply
I agree my counterexample isn't the prototypical one, and I even agree that if the intention is to teach someone who understands single-variable continuity but not multi-variable continuity, we ought to include such prototypical examples. It's just been so long since I've taken (or even since I've taught) multivariable calculus, I had forgotten all about them.
However, I am unapologetic about my example. I designed it specifically so that x → 0 + lim f ( x , k x ) = { ∞ 0 1 / 2 < k < ∞ otherwise has only those two possibilities. This is more interesting to me because it suggests a method for creating a function f ∗ whose restriction to a line is discontinuous for exactly one line and the discontinuity is essential.
As for your last comment, of course f ( 0 , 0 ) = 0 and f ( x , y ) = y 2 + x 4 x 2 y for all other points would work.
Problem Loading...
Note Loading...
Set Loading...
Consider the functions g ( x ) = ⎩ ⎪ ⎨ ⎪ ⎧ x 2 − x 0 0 ≤ x ≤ 1 1 ≤ x ≤ 2 otherwise and f ( x , y ) = { y g ( x / y ) 0 x , y > 0 otherwise
Then we claim that this f satisfies the two conditions but is not continuous at the origin.
Property 1: Fix some X . If X ≤ 0 then f ( X , y ) = 0 for all y , which is continuous. Therefore, let's assume X > 0 . Then f ( X , y ) = { g ( X / y ) / y 0 y > 0 otherwise = ⎩ ⎪ ⎪ ⎨ ⎪ ⎪ ⎧ y 2 X y 2 − y X 0 0 ≤ y X ≤ 1 1 ≤ y X ≤ 2 otherwise = ⎩ ⎪ ⎨ ⎪ ⎧ y 2 X y 2 − y 2 X 0 y ≥ X 2 X ≤ y ≤ X otherwise which is continuous (note that the pieces "fit" together and at the only possible problem point of y = 0 , the function is locally zero, so continuous)
Property 2: Fix some Y . As before, if Y ≤ 0 , then f ( x , Y ) = 0 for all x , which is continuous, so we may assume Y > 0 . Then f ( x , Y ) = g ( x / Y ) / Y = ( h ∘ g ∘ h ) ( x ) is just the composition of the continuous function g and the continuous function h ( x ) = x / Y , hence continuous.
f is not continuous at the origin: To see this, note that f ( 0 , 0 ) = 0 x → 0 + lim f ( x , x ) = x → 0 + lim x g ( x / x ) = x → 0 + lim x 1 = + ∞ .