That "a-ray" seems pretty damning. My guess is that by the time he wrote his article, Masefield had forgotten - or had never fully recollected - the words he'd heard sung, so he turned to L. A. Smith.

Some of his lyrics also seem to be from "The Lowlands of Holland." I personally doubt that Masefield would have consulted two separate ballads to fake his lyrics. Why bother? He wrote actual poetry himself.

"Drowned in the windy lowlands" is pretty clearly a poetic revision of "drowned in the lowland sea" or something of that sort. It also sounds more like Masefield's poetic style than anything remotely traditional.

The Davis & Tozer problem is a mystery, but I can't imagine that Masefield created it intentionally. He may well have used D&T's melodies & arrangements for precisely the reason you suggest - or else his publishers decided that melodies were needed and D&T was handy.

I don't know anything about Masefield's own musical abilities.

While a female narrator in a song like "Lowlands," sung by men at work, would probably be impossible today, Masefield may have picked up the outline of his version from somebody who'd learned it thirty years before M. went to sea - in other words, the 1860s, near the height of sentimental Victorianism. It might not have seemed so peculiar then.

It seems perfectly reasonable to me that a shanty with a refrain of "Lowlands, lowlands...!" would borrow lyrics from two well-known broadside ballads about lowlands. I'll even suggest that the form of Masefield's shanty (though not every word of it) may well represent the original pattern. The dead lover could easily have been changed to a female later on. I don't have Hugill handy so I can't tell what his corresponding version is like.

But the "a-ray" remains difficult to explain.

The line "I'm afraid you're a bad one, kind sir, she said," is absolutely believable, esp. if "bad one" was pronounced "bad 'un," a fairly common 19th C. phrase.