Hi there. I'm investigating using matplotlib for plotting of Adaptive Mesh Refinement ( http://en.wikipedia.org/wiki/Adaptive_mesh_refinement ) data -- the primary characteristic of which is that it is of non-equal resolution. I've used the scipy/delaunay method, as mentioned in the cookbook, but unfortunately that provides a level of interpolation that is not always desirable; very often when plotting data, we want to be able to see clear cell boundaries, as well as boundaries between resolution levels. Essentially what I have are the following pieces of data: x, y, dx, dy, z. The simplest way is for me to sample this using a loop over points in the module where I handle the data; however, what I'd like to be able to do is hand it off to matplotlib, and the on-the-fly change the x,y (and z) bounds. (This seems as though it would be the more efficient manner of handling the data, anyway.) Is there a way to do this? If not, would it be terribly difficult for me to implement? I've browsed the code, and it seems that the best starting place would by pcolor in lib/matplotlib/pylab.py or src/_image.cpp. I very much would like to leverage the abilities of matplotlib -- specifically, I'm very excited about being able to plot this data, and then overplot contour or quiver plots (which I have done with my data using the delaunay method.) Any ideas? Thanks! -Matt