Hmmm. Do they, Alasdair? One could argue that, for a given expenditure of resources (primarily petal area), the flower would be brightest if it was a complete disc. Feathering the edges of the petals may allow a greater diameter, but the overall brightness (ie. from a distance) would be diluted by the (presumably darker) background.
I would be interesting in seeing pictures of such cut-petalled flowers (a selection of specis) under UV lighting; maybe there are 'honey guides' extending out to join up with the cuts?