I've been doing some research into the BARNBY family of York, initially using Find My Past's collection of York Parish Registers, and have found that in many cases it is the Bishop's Transcript images and their FMP transcriptions that have been added to the site, when the Parish Registers will almost certainly contain more information. I first realised this when I cross-checked on Ancestry for the baptism of Samuel BARNBY, on 09 August 1785 at St Sampson's, and found two transcription entries. The first, from a collection entitled "England, Select Births and Christenings 1538-1975" gave his birth and baptism dates plus his father's fore- and surname, and his mother's forename. This information is identical to that on FMP. The second, from a collection entitled "England & Wales Christening Records 1530-1906" gave the same information, plus his mother's maiden name, and the names of both paternal grandfathers. The latter is obviously invaluable when trying to confirm that you have found the correct ancestors for someone. I'm assuming that this is a transcription of the PR, so I now need to find and check a microfilm or digital copy of the original register. In other instances, I've found that the transcription omits vital information that is clearly visible on the original register image. It's obvious, but I'd recommend always to check the original Parish Register, if it still exists, rather than relying on transcriptions or BTs, both of which are prone to transcription errors or just omitting information, as the process of producing an Ancestry or FMP transcription involves a number of stages, at any of which errors/omissions can be introduced: 1. The clergyman writes information in the Parish Register, relying upon often illiterate parents, and sometimes without the diligence that might be expected 2. At the year end, he or another parish officer has the boring, administrative job of compiling the Bishop's Transcripts, so he may decide to copy just the minimum information needed, without necessarily being too bothered about accuracy, and also by possibly having to interpret someone else's scrawled handwriting 3. Years later, another person tries to read the BT (more scrawled handwriting, and possibly a microfilm copy of dubious quality!), and copy the information into a format suitable for being printed. This may be typed, or be yet another manuscript 4. The typesetter at the printers prepares the blocks ready for printing, and a proof of this is checked for accuracy, but maybe by yet another person 5. Ancestry, FMP etc. type their transcript (or maybe use an Optical Character Recognition program) into a computer to put online, possibly using the PR or BT, but possibly instead using the printed transcript 6. This information is put online to be lapped up by a sometimes un-sceptical audience Someone could probably work out the statistical probability of the information at step 6 being identical to that at step 1, but I bet it is low enough not to use it as the basis for, potentially, years of wasted effort researching the wrong family, for wont of reading the original parish registers. Don't get me wrong, I'm a great fan of the online information provided by Ancestry, FMP etc., as it saves a lot of time and effort if used correctly. But, as was said of the IGI years ago, it is best used as a finding aid, and cannot be relied upon to be accurate.