<<snipped>> Whether or not the parish appears in the search results depends on the record set. For example, if you search 'Kent baptisms' the parish isn't shown in the search results, but if you search 'Canterbury baptisms' then it is. I would guess it's to do with how the data is formatted (the transcriptions for these two record sets use completely different templates). <<snipped>> I agree, Peter. "Diocese of Chester parish registers of baptisms c1538-1910" shows "Cheshire, England" for the person I'm looking at. The "Diocese of Chester Bishop's Transcripts of Baptisms c1600-1910" shows "Davenham, Cheshire, England" for a similar event (but in the different data source). The "transcriptions" (indexes for goodness sake!) show different fields have been populated. Now, these Cheshire records were indexed by LDS and I have said on several occasions, that the LDS indexing was nothing like consistent across the Cheshire collection. I suspect that when FMP loaded the original Cheshire Collection, it did data quality checking and mapped the loading data to its (old) database in such a way that it always got the parish name in the right spot. (At least I *think* it did!). Loading all their old databases into the new databases, I think the quality checking has been skimped - perhaps on the basis that it loaded once, so surely it was OK? Or perhaps it was simply the volume of data to be loaded that made it impractical to check to the same level? Whatever the reason, we appear to have a series of data quality issues that have impacts varying from minor irritations to major fiascos (e.g. the issues with searching a census by birthplace). When I can summon up the enthusiasm I will try to compare like with like to see what causes the loss.... Adrian B