R help - Feb 2008 - scatter plot with point colors from kernel density

I am using R 2.6.1 on a 64 bit PC running Windows XP Pro.  

My main problem is that I want to produce a scatter plot with density
estimation, where the colors of points correspond to a kernel density estimate
at the location of the points.  I have not found a package that will do this.

If there is no package, one option would to estimate the bivariate density
first, and then use it to define a color table for the points.

If I obtain a bivariate kernel density for X and Y coordinates as follows:  


 require(MASS)  

 x.coord <- rnorm(1000)
 y.coord <- rnorm(1000)  

 kernel <- kde2d(x.coord, y.coord, n = 100)  

par(mfrow=c(2, 1))  

 image(kernel)  

 points(x.coord, y.coord)  

 plot(x.coord, y.coord, main="I want to color these points using the kernel
density")


how can I extract the kernel density estimates at the x.coord and y.coord
locations for each point?
What I would like is a matrix with three columns for "X",
"Y" and "DENSITY", like this:
> matrix  
      X      Y      DENSITY
1    1.56  -0.03     0.05  

 ...  

Is there an easier way to do this?  

Thanks in advance,  

Jim
  
	[[alternative HTML version deleted]]