A really nice scene, Bob. The lithops also came out well. The only problem I now see with the cluster is that some specimen are floating. With less rugged ground and rotation to ground applied (use an extra compute normal with a small patch size (0.5m or so) at the end of the chain), you might partially solve it. And/or slightly sink them into ground (object based).
Best would be if we had... can't find the words... ah, collision detection, and use single objects. Later maybe.