Thanks, I just had a look at that 1901 census return as well. As you say, Hoddam.
So the facts we seem to have are:
- The James in the 1901 census is married to Margaret, living in Gretna and giving his birthplace as Hoddam.
- The marriage you have to Margaret Chalmers shows William as James's father.
- There is a birth for a James Johnston in Hoddam in 1859, but he is not the son of William and Janet, but the son of David and Mary.
Logic says we are looking at 2 different James Johnstons here, and to complicate the issue there is no guarantee that Johnston / Johnstone / Johnson is always spelled the same way in the records over the years, depends entirely on who gave the information and who wrote it down.
To separate the strands, because I'm sure it will be possible to find 'your' James, can you start at the last definite thing you had? Who is James, in that is he your gr.grandfather or similar and what is the oldest absolute thing you know, like the year of his death, or the birth of his descendants?
I think we need to start from the oldest definite, then go backwards one step at a time because whilst I can see that the 1901 census James can be the boy born at Hoddom, I don't think the James who married in 1879 can be, unless he swapped fathers somewhere along the way.
Conversely, it's perfectly possible that 1901 James and 1879 James are the same chap, but then that would rule out the 1859 birth...
Maddening, but then that's half the fun!