# Proving Limits

1. May 22, 2009

### roam

1. The problem statement, all variables and given/known data

http://img40.imageshack.us/img40/39/20688555.gif [Broken]

a) Show that if C is any straight line through (0,0) then $$\lim_{(x,y) \to (0,0)}$$ along C exists and equals 1.

b) Show that the limit as (x,y) -> 0 doesn't exist.

2. Relevant equations

3. The attempt at a solution

I really need help with this question! I know that x2 is a parabola and any straight line through the origin either intersects the parabola at some point and remains above it until 0 is reached (or lies on y=0, in which case $$y \leq 0$$ and $$f(x,y) = 1$$) but still I don't know how to "prove" part a).

Last edited by a moderator: May 4, 2017
2. May 22, 2009

### tiny-tim

But you've done it!

If, sufficiently close to the origin, it remains above the parabola until 0 is reached, then the value along that part of the line is 1, so the limit exists …

what's worrying you about that?

3. May 22, 2009

### roam

Hi!

In part b) asks to 'show' that the limit as (x,y) -> 0 doesn't exist...

4. May 22, 2009

### tiny-tim

Well, it obviously doesn't exist …

to prove it, use the definition of limit (y'know, the epsilon and delta thing)

5. May 22, 2009

### roam

Yes I see. I know how the epsilon & delta proof works but I don't know how to exactly apply it to this problem. I appreciate it if you could help me get started on this, thanks.

6. May 23, 2009

### tiny-tim

Well, if the limit exists, and is 1, say, then that means that given any epsilon > 0, we can find a ∂ > 0 such that √(x2 + y2) < ∂ means that |f(x,y) - 1| < epsilon.

So ty it with epsilon = 1/2

7. May 23, 2009

### HallsofIvy

Staff Emeritus
Simpler, I think: try approaching (0,0) along the curve y= (1/2)x2. In that case, y is always less than x2 so f(x,y)= 0.

8. May 27, 2009

### roam

You mean $$\epsilon = \frac{\delta}{2}$$?

9. May 27, 2009

### Staff: Mentor

No, he meant epsilon = 1/2.

10. May 27, 2009

### HallsofIvy

Staff Emeritus
Using an $\epsilon- \delta$ proof, $\epsilon$ is "given" and $\delta$ depends on $\epsilon$, not the other way around.

But I still think my suggestion is simpler. I'll just go off and sulk!