I agree that canoeing can often be traced to 'Paper Butterfly' (PB) ancestry but one can ask is that due to some general genes in PB that cause canoeing or is it perhaps caused by a specific gene for pigment or pattern effect in PB's flowers. In other words is the canoeing linked specifically to a pigment or pattern?
One way that one might get a few clues to help decide which type of genetic effect might be causing the canoeing is to look at the offspring (that have photos available) of crosses directly involving PB as the pod or pollen parent with the other parent being a registered cultivar. From the AHS database: plants such as 'Butterfly Cravat' from ('Gay Cravat' x 'Paper Butterfly'), 'Butterfly Eye' from ('Forever in Time' x 'Paper Butterfly'), 'Carolina on my Mind' from ('Isle of Zanzibar x 'Paper Butterfly'), 'Cecile Holcombe Black' from ('Paper Butterfly' x 'Sherry Lane Carr') and so on. One needs to try and choose crosses in which the other parent does not apparently have canoeing or similar patterning as PB, which may be difficult sometimes.
With a large group of such offspring, then separate the offspring into two groups, those that show canoeing (in their photos) and those that do not. Are the flowers of the two groups the same or are there differences between groups? If they are the same then it is likely that the genetic effect is general but if they differ, especially if the group showing canoeing has patterning that is different or flower colours that are different from the group that does not show canoeing then it is likely that the genetic effect is directly linked with the pattern or the pigment or both.
