salto
Sorry Rua, but I disagree. Although saltus indeed is an inflected form of salto, approving sault under descandants of salto sets a dangerous prerequisite. For instance, I could easily add the following Romanian words as descendants, because hey, they're inevitably from the same source right: salt, sălta, săltare, săltăreț, săltător, săltătură, saltație, etc. I we apply the same logic to the other Romance languages the result would be a descendants section overcrowded with terms indirectly derived from the lemma entry. Do you see what I mean?
I seem to recall this being discussed, but I don't believe we reached an agreement. But do you see what I mean about overcrowding the descendants section?
Robbie SWE, do you think you could weigh in here?