By the way, I have found some background information on Bronson and his methodology via JSTOR. It would appear that he used some kind of statistical techniques to try to identify tune families - though I have not yet been able to access the full article, being up to the limit on my monthly allowance. I suppose that today he might have used computer techniques to do this.

For anybody interested the reference is

Some Observations about Melodic Variation in British-American Folk Tunes
Bertrand Harris Bronson
Journal of the American Musicological Society
Vol. 3, No. 2 (Summer, 1950), pp. 120-134