A most interesting survey of pollen fertility in narcissus hybrids .I like to make two remarks. 1. From several wild narcissi tetraploids are available (Zonneveld (2008) Pl syst Evol 275:109-132) . It would be very worthwhile to seek these out. This would circumvent the low fertility of the triploids 2. There is another, I think simpler way, to investigate the pollen. Take some pollen on a slide and add a drop of cottonblue lactophenol . Within minutes good pollen is dark blue, bad pollen stays colourless as can be seen in a simple microscope. I have used that for many genera.