And you have to bear in mind that those "sources" were not original sources - they were just the people/settings from where the song was picked up at the time that collection was compiled (1927, I've just looked it up). Each of those in turn may have other sources - someone could learn a song at their grandma's knee, but Granny herself could have learned it from a printed broadsheet. And the middle-class contributors could have learned their songs in a number of ways - printed sources, oral sources such as other singers of various backgrounds, also radio, gramphone records, etc.

Any song collection is just a snapshot of what was known or being sung at a particular point in time. It would be quite interesting to do a study now of, say, the songs sung at a folk gathering such as a festival, and ask each singer where they got the song from. Some will have been acquired orally (many singers still don't read music) from other singers either in person or in recordings; some will have been learned from books; some from websites like this one; and some will be a compilation of several of these. Oh, and some self-penned.

Of course, nowadays the more scholarly folk singers will be meticulous about their attributions, and explain that they're singing a version collected from a certain singer by Cecil Sharp in 1908, but that only takes it back as far as the days of the early collectors - the origins of many of the songs will never be known.

Marje