I've been doing some research over the past few months that makes extensive use of the new indices recently published by the GRO of their birth and death registers for England and Wales. There are techniques available now that were not available without them.
There have been online indices of the England and Wales BMD civil registers for as long as I have been doing genealogy (notably at FreeBMD but also from most of the commercial online genealogy providers with coverage in the UK) but they all have problems arising from the chain of events that led to their production.
A brief outline of the process that led to the compilation of the original indices from 1837 onwards may help understand the issues:
- An event (birth, death, marriage) took place and the details was recorded by a Local Registrar (or in the case of religious marriages, by a minister or other celebrant who provided the information to the Registrar).
- These original details were kept locally, but every quarter a 'True Copy' of all the entries was sent to the General Registrar (via a Superintendent Registrar who had overall responsibility for a Registration District).
- The Superintendent Registrar also indexed the events in his Registration Distict, which could be searched for a fee.
- The General Registrar (or rather their Office) retained the copies provided by the Superintendent Registrar and also indexed them to facilitate searches. (The information held in the indices varied with time -- mother's maiden name and age at death, for example, were not recorded until 1911 and 1866 respectively.)
- FreeBMD and other organisations used the paper indices produced by the GRO to create online browsable or searchable indices.
Of course, nothing that is done now can addresss problems introduced at the begining of this process: events that weren't registered at all, or were registered with misleading or mistaken information; and the likelihood that information was lost or corrupted in the chains of copies and indices that were made. (See "A Comedy of Errors" or The Marriage Records of England and Wales 1837 - 1899 by Michael Whitfield Foster if you want more detail on the process and the errors introduced; as far as I know, it's out of print now but you may be able to find a used copy.)
However the GRO has made some improvements in its own indices, which are based on work done for an (incomplete) project to digitise their records. For births between 1837 and 1916, and deaths between 1837 and 1957, they have gone back to the copies they hold of the original register entries and created new indices that are consistent in the information they provide: age at death for ALL deaths, and mother's maiden name for ALL births.
Inevitably, new errors and discrepancies have been introduced: there are entries in the original paper-based indices that don't exist in the new online indices (perhaps because the events were missed in the new indexing process, perhaps because they were misindexed either now or when the paper indices were produced, or perhaps because the copy entries of the original registration have vanished between production of the two indices.) There are also entries in the new indices that didn't exist in the old ones, and there have always been discrepances between the local indices and the central ones.
Even with the problems that still exist the new indices provide the opportunity to do research on families that was previouly impossible (or hideously expensive if you decided to order multiple certificates to solve a problem). You still have to correlate the information you find with other sources of information to confirm (or otherwise) the reliability of your conclusions, but it's a big step forward.
- If you know the mother's maiden name for one child in a family, you can search the indices for other children with the same surname and mother's maiden name. This can throw up hitherto unknown children who were born and died between censuses, and allow you to complete a picture of that family. It's best to cross-reference the results from the indices with census and/or baptism records to make sure you're dealing with the right family, especially where the surname of one or more parents is a common one; and to confirm that there's a matching death record for the short-lived child.
This approach can also help you detect unexpected patterns of movement in the family, if you find a child who was born and died in a location where you didn't know your family had lived -- but make sure there wasn't another 'matching family' in that location.
- If you're trying to track down a death between 1837 and 1866, having the age at death can help identify the correct individual (as it always has done after 1866).
However, if you're dealing with infant deaths (shown as 0 in the old indices) there's a new way of narrowing down the possibilities. In the new indices, age at death was supposed to be rounded down to the nearest year, but that hasn't always been done, so some deaths will show (for example) an age of 5 when it's recorded as 5 months. Cross check the age at death (after 1866) in the new indices with the age in the old indices. If the old indices show a zero, and the new indices show an age up to about 12, it's a fair bet that the new indices have recorded a age in months as an age in years.
- Even though there isn't a new index of marriage records, there's a way to confirm marriage partners using the new birth indices in conjunction with census records. The existing marriage indices (searched online) will often give you multiple names with the same reference number, and no clue as to how to sort those names into couples. Parish records might help, if the relevant records are available on line. If not, search the birth indices for 4 or 5 years after the marriage for children with each combination of male surname and female maiden name (I narrow it down initially to registration districts close to the marriage, but have had to spread the net wider on occasion). For each child found, search for it in the census records -- you're looking for one or more children with the right name and age, with the same surname and mother's maiden name and the corresponding combination of parents' forenames. Of course, this doesn't work for women who weren't of child-bearing age when they married, or who emigrated early in the marriage, but it's still a useful technique.
A resource I haven't mentioned yet which may not be well known to everybody is the LocalBMD indices. There are being produced by local volunteers from the original records held in Registration districts. Coverage might be incomplete in some districts or totally lacking in others, and the search facilities and information indexed differ from district to district, but these indices have the advantages that (a) they're produced by volunteers with local knowledge and (b) they're being made from records that are closer to the originals in the chain. Go to UKBMD to understand what's available and how to access it (but be aware that GRO index references don't apply to the local records and vice versa).