Well, what you give in the first response is NOT a definition of "differentiable". It certainly is not what is given in the Wikipedia site. I think you must have copied wrong- you have an "f(x,y)" that doesn't belong in there.
Essentially, what Wikipedia says is that f(x,y) is differentiable at [itex](x_0,y_0)[/itex] if and only if there exist a vector [itex]u\vec{i}+ v\vec{j}[/itex] and a function [itex]\epsilon(x,y)[/itex] such that for for some [itex]f(x,y)= f(x_0,y_0)+ (u\vec{i}+ v\vec{j})\cdot ((x-x_0)\vec{i}+ (y- y_0)\vec{j})+ \epsilon(x,y)[/itex] and that [itex]\lim_{(x,y)\to (x_0,y_0)} \epsilon(x,y)/\sqrt{(x-x_0)^2+ (y-y_0)^2}= 0[/itex].
In that case, the vector [itex]u\vec{i}+ v\vec{j}[/itex] is the "gradient" of f at [itex](x_0,y_0)[/itex].
So I would recommend that you go ahead and find the gradient of f(x,y)= xy at [itex](x_0, y_0)[/itex] and use that to determine what [itex]\epsilon(x,y)[/itex] must be. If you can then prove that [itex]\epsilon(x,y)/\sqrt{(x-x_0)^2+ (y- y_0)^2}[/itex] goes to 0 as (x,y) goes to [itex](x_0,y_0)[/itex], you are done.