knn classifier with own distance function

Question

I am using a Knn classifier in MATLAB. Here is my Code:

load fisheriris
x = meas(:,3:4);

newpoint = [5 1.45];

[n,d] = knnsearch(x,newpoint,'k',10, 'Distance', 'euclidean');

Now I would like to use my own distance function. To be sure, that it is working, I first would like to define my "own" Euclidean distance. After that I would like to try the ChiSquare Distance.

My problem: How can I create my own distance function. I know I should use pdist(), but I am not able to get it working. Would be great to get some hints.