I am working on an online reference of digital micro scans of droserae seed coats in mutual research with the Senior Seed Analyst at the Idaho State Herbarium. You are right, for many species the seed is very diagnostic! An electron microscope isn't needed to see the seed details, just a good 25x lens or one of those microscopes that come in hobby packages for kids. The bad news is that for some of the South African species, the seed testa details are alike for several species (natalensis/aliciae) (nidiformis/dielsiana). Generally though seed coats are very useful: they can save a grower from wasting time on falsely identified species. The ID problem is *huge* regarding this genus, and I hope my reference will be useful to hobbiests and commercial dealers. I hope in time to expand the project to also include digital scans of the stipules, flowers and scapes which are also often used to identify the species. My challenge is to locate and confirm the ID of the seed that is being scanned by my associate. This pretty much means finding people in various countries where the plants are native that can access the populations, and provide collection data along with the needed seed. It's slow going - so far we have recorded some 20 species, and there is seed waiting to be scanned. There has been a wonderful response from most of the CP community for the work.