I do not have a fool proof answer to this but I do have some suggestions. If anyone has a better idea please let me know because I am very much in the same boat.

1. In the description column of Griffith's look for mention of a house (you will need the full Griffith's not the CD ROM), a house in one townland but not the other may indicate the same owner, but please remember this is far from conclusive.

2. Check the cancelled books, these cover the period from about 1860 until the 1940s - 50s. Each change of land holder is noted and dated, two people with the same name but changing on different dates are probably different people, two people with the same name changing on the same date and succeeded by the same person are almost sure to be the same person. The weakness of this method occurs when succeeding generations have the same name, that makes it almost impossible.

3. A check of Guy's directory is often worthwhile, unlike most directories Guy's includes the principal landholders, each individual will only be listed once, ie two people with the same name = two different people. The weakness with this is that Guy's only starts in the 1870s.

Often the real problem is that people who were related and had the same name lived in the same locality or even the same house, you only have to look at the 1901 census to realise the extent of the problem.