[MUSIC PLAYING]

Hello, everyone. My name is Audrey Collins, and I work at the National Archives of the UK, and I'm the family history specialist there, which is the best job in the UK Civil Service, I think. I'm here today to talk about the civil registration indexes for England and Wales. That's the indexes to births, marriages, and deaths in England and Wales from July, 1837 to the present day. And I'm talking about the indexes specifically rather than the records themselves for a very good reason, which I will come to. Now, the registers in the indexes--they are not held by the National Archives where I work. They belong to the General Register Office--or GRO--which is a completely separate government department. Now, I've been researching family history for about 30 years, and I've been particularly researching into the background and the origins and the ways of working of the General Register Office for about 20 years, and this is what started me off. In 1999, a man called Mike Foster, who was based in New Zealand, but was originally from England--he'd been researching his family history, and, like many people with English and Welsh ancestors, he had come to the conclusion that the indexes--particularly the indexes to marriages--were very deficient. And on a trip to England once, he managed to get absolutely unprecedented index access to the records in the General Register Office in its headquarters at Southport in Lancashire. Now, the reason that I want to talk about the indexes is because they are the only way that we can get to the records. With most records, if you don't trust an index, you can do it the hard way. You can go and look through a parish register or a census or many other kinds of records. But for these civil registers, the only way to get to them is through the indexes, and the reason for this is that, in 1836, when the act was passed setting it all up, there is an unfortunate phrase in the act, which was that the information could only be given in the form of certified copies. And that is why they are expensive to get at, and you can't go and look at the records. The law doesn't allow it. So that is why the indexes have an importance that other indexes, which are secondary sources, don't have. Now, I got this book, and Mike Foster came to some very interesting conclusions. By looking at the records that they hold in the General Register Office, which are called quarterly returns--that is, every quarter, that's every three months--every registrar in England and Wales would have to send copies of every event that they'd registered, and they were kept centrally at Somerset House and turned into the registers and indexes. And those are the things that we now access when we're using the national indexes. Now, there are also local indexes, which I'm not really going to talk about, but I am going to mention them briefly because they are important, and very valuable. But Mike Foster's work in the registers and the indexes in the GRO--he was able to prove conclusively that there were omissions, that there were places where it was obvious that somebody had turned over two pages at once when indexes were being microfiched or sometimes that someone had turned over two pages at once when the quarterly return volumes were being microfilmed by the General Register Office itself. So he came to a lot of interesting conclusions, and he was able to prove a lot of things that we had often thought. He also raised a lot of questions because what he was trying to do was to work out how these indexes were produced, and, like me, he believed that understanding the way a record was created and who created it and the mechanism by which they did it--that can be very important in helping you to understand the records and to make the most of them, and also to find strategies when maybe you don't find the result that you want. So it was reading his book--and my copy is covered in scribbly little pencil notes--that inspired me to go and do the research to find the answers to the questions that he wasn't able to answer. He made some guesses, and they were quite good guesses. Unfortunately, a lot of them were wrong, but that's not his fault because the information that you needed to answer those questions was not in the General Register Office. It's a government department. It has the records and everything it needs to do its current job, which is registering events and producing certified copies. Like any other government department, it doesn't have all its own records of its own history. Those records are held in the National Archives. And I started researching this even before I worked there myself, so I was there just as an ordinary reader, and I spent a very long time chipping away at this research, and I'm still doing it, time permitting. And more than a decade later, after Mike Foster's book, I was very pleased to be the co-author of a book called Birth, Marriage, & Death Records. Now, this is about births, marriages, and deaths of all kinds in the whole of the British Isles, so it's not just about civil registers, but I did write the three chapters in it on English and Welsh civil registration. And my co-author and so-called friend, Dave Annal, was introducing me once when I was doing a talk very much like this, and he said, "What Audrey doesn't know about early civil registration isn't worth knowing." And then he paused, and he said, "And quite a lot of what she does know isn't worth knowing either." Oh, I forgave him because it was such a good line, and he's probably right. So that was me. That was how I started, and that's how I was able to write up some of my research findings, but you can't fit it all into three chapters. Anyway, I want to try to explain to you how the system worked. So I've got a diagram here of the way the registration system was set up. Now, you see at the top that there is Somerset House, and that was the headquarters of the GRO until 1974. And even today, they still get post addressed to Somerset House, London. And I think it still finds them. Now, when civil registration was set up in 1837, the whole of England and Wales was divided into registration districts. Now, these were based on the poor law unions, which had been set up about three years earlier, and you will find that sometimes the term union is used to mean registration district even by the General Register Office itself. And it makes a lot of sense. Although it's not strictly accurate, it's close enough. And the word union has got only two syllables and five letters whereas registration district is much more of a mouthful. So union is OK, so you will often see it in the context of registration districts. Now, in charge of every registration district, there was a superintendent registrar. He didn't do most of the day-to-day registering. He had to countersign things and he had lots of other duties, but the actual registration business was done principally by registrars of births and deaths. Each district was divided into subdistricts, and the registrar of births and deaths was responsible for registering all of those events that occurred in his own district. And until the 1870s, it was always a he. They rather carelessly didn't make it a male-only occupation in the legislation, which happened with many other things, so there was never any--women were never forbidden to be registrars, but the first ones began to be appointed in the 1870s. In this year, where we're celebrating the centenary of women's suffrage, I just thought that was a nice thing to put in. So the superintendent registrar was in charge. The registrars of births and deaths registered everything in their own subdistrict. Now, on the right-hand side of the diagram, you'll see there were registrars of marriages. These were responsible directly to the superintendent registrar, and some of these were also registrars of births and deaths, but they didn't have to be. And these people were responsible for performing civil marriages anywhere in the registration district. Now, in practice, a true civil marriage in a register office was extremely rare. Until well into the 20th century, the great majority of marriages took place in churches and principally in the Church of England. The registrars of marriage would perform some civil ceremonies, but mainly, what they were responsible for was performing marriages in places of worship which were not the Church of England. Historically, the Church of England clergy have always been able to perform marriages. Just by virtue of being an ordained minister, that gives you the right to perform marriages. And they had had a monopoly on this from 1754 up until 1837. The only exceptions to this were Jews and Quakers. And as a proportion of the population, these were very, very few. So most marriages were performed by Church of England clergy. And on the diagram there, you'll see--where where I put clergy, I mean Church of England clergy--and the line to that little picture is a dotted line, and that is because the Church of England clergy, although they had the historic right to perform marriages, and they had obligations under the Civil Registration Acts, they were not directly responsible to the superintendent registrar or the registrar general. There was no chain of command. The registration authorities could not tell the clergy what to do. I have some--well, they're not really instruction manuals. They could issue instructions to the registrars, but what the clergy got was "suggestions for the guidance of." Because their monopoly was really being broken by the introduction of civil marriage and civil registration generally, the Church of England wasn't 100% cooperative, particularly to start with. There was only a very few clergy who were really, really difficult, but an awful lot of them were just not terribly good at paperwork. And even today, most superintendent registrars will tell you that, in their district, there are usually one or two clergymen that they have to--it's very difficult to get the paperwork out of them on time. They're busy. They're overworked. And they're just not very good at it. So they're not necessarily being uncooperative. And the other dotted line leads to authorized persons, and these were people who, from 1899, where there was an amendment to the act--this meant that, for the first time, people who were typically Baptist, Methodist ministers, Catholic priests could apply to become authorized persons. That means that they could perform marriages in their own places of worship without having to get a registrar in to do it. So from 1899 onwards, if you have a marriage in a non-Church of England place of worship, then you might find it's performed by an authorized person or an AP, as they're called. And I've put "authorized persons plus" with great apologies to anyone who is of a Jewish or Quaker ancestry because of--I've just reduced them to a plus. But as I said, as a proportion of the population and a proportion of the marriages that were registered, that is absolutely minute. So every quarter, all these people--all these registrars, and all these clergy, and, later, the authorized persons--they had to make copies of every event that they'd registered and send those to the superintendent registrar, who would forward them on to Somerset House, and that's where they would be--so this is why they were called the quarterly returns because they did it every quarter. And they were bound into volumes, and they were indexed. These are the indexes that we use. Now we use them online, but we used to use them in big paper or parchment books. And the registers themselves--the the quarterly return volumes--that is what the certificates were produced from. And when I said that a union wasn't strictly accurate for a registration district, a certificate is not, strictly speaking, what we're getting. When you order something--we all say we're ordering a certificate--what we're actually ordering is a certified copy of an entry in the register. But life is too short to say that all the time, so I'm good with certificates.

This didn't always go terribly smoothly. And this--I found this in the registrar general's correspondence--this unfortunate incident. "In May last, the clerk sent by the superintendent registrar of Leeds, Mr. Lampen, had willfully destroyed the packet of quarterly returns instead of posting them to the GRO, consisting of 2,086 entries in all. The registrar general instructed the superintendent registrar to obtain duplicate copies as quickly as possible at a cost of 15 pounds, one shilling, and fourpence." And the registrar does seem to have done that because there do seem to be about the right number of entries in the indexes, but he must have had an awful sinking feeling when he had to do that, and I have no idea why the clerk destroyed them, and I doubt if we'll ever find out. But we can make up our own stories. But it was the returns from the clergy, for reasons that Mike Foster was able to demonstrate, caused more problems. And then, in 1899, when you started getting the APs--"The certified copies of marriages since April last by about 1,000"--1,000, that's a whole lot of new people--"authorized persons are now being received in the office, and they naturally display an infinite variety of blunders and contraventions of the rules and regulations of a more or less serious character." Now, I love the "naturally" because this is what they expected. You could assume--sigh. I think it's a polite way of saying, "If you thought the Church of England clergy were bad, you should see what we're getting now." So this is just another reminder that the marriage indexes have all sorts of reasons to be more deficient than births and deaths. Now, when they got to the General Register Office, this is the process, which I've discovered from a lot of documents that we hold, of how they got from being a register entry to being an index entry in the GRO. And I confess that that is not a GRO clerk. It's a lawyer's clerk, but he's from about the right period, the 1830s, when registration started, and I like the picture. First of all, they would go to the examining room, and this is where a whole lot of clerks would look at these quarterly returns, and they would look for obvious errors. Now, they weren't mind readers. They couldn't tell if a name was wrong, but they could spot things which had not been filled in correctly or where something like the date of registration of birth was earlier than the date of birth--obvious things like that. So anything that was problematic, they would query that. There was a whole correspondence section would deal with the superintendent registrar, and this would go back and forth, and hopefully they would resolve all these queries. They didn't always, obviously, because life's not like that, but this was their job. Then the quarterly returns--these pages would go to the arranging and paging section. And each page--first of all, for births and deaths, they would be up to 10 entries on the page, which is exactly the same as in the registrar's own original registers, and for marriages, there would be--first of all, there would be up to four marriages on a page. In the early 1850s, this was reduced to two.

And they would look at all these pages, and then they would sort them into order. Now, within each subdistrict, you would get all your births and all your deaths. And for all the marriages within a whole district, they would be put in order. Now, within a registration district, all the births and deaths would be bundled together in subdistrict order, usually alphabetically, and the marriages would be alphabetically by the church that they had come from, and any registrar marriages and authorized person marriages--they would be at the back. They would all be bundled together in subdistricts and then in districts, and then the districts were assembled into volumes. Now, a volume is a completely artificial geographical description. There wasn't an office in charge of a volume, but it's a geographical area. And first of all, these were numbered. There were Roman numerals, starting with one in London and working out around the country from the southeast and so forth and up to the north, ending in North Wales. So you had Roman numerals for the volumes up to the early 1850s, and then they changed to numbers and letters after that. And then there was another reorganization in 1946, but the principle was still the same--that you have subdistricts, districts, volumes. And once they were assembled into volumes, then they were given page numbers. So all the entries, all the pages from a particular district--in fact, from a particular volume--would be numbered. So starting with the first page of the first subdistrict of the first district right through to the last subdistrict of the last one. So over the years, if you see lots of entries from the same district--I mean, if your family stayed in the same place for a long time--you will often find that all the births or the deaths--they will have fairly similar numbers because they would have roughly the same range of numbers year after year. So they were arranged in pages. And now you've got some big volumes, and then it was time to index them. Now, the first thing you had to do is to transcribe. So the clerks, who were called transcribers--what a surprise--they would go and extract the information from each entry that was going to go in the indexes. Now, to start with, no matter what the entry was, whether it was a birth, marriage, or a death, they would extract the last name, the forenames, if any--you do get some entries that don't have a name at all--and then the district, and the volume, and the page. And that would be on a transcription slip. In fact, it would be on a big transcription form which was then cut up into slips. And then--you guessed it--they'd be sorted. And a whole other army of clerks had great big piles of paper slips to sort into alphabetical order, first of all, by initial letter of surname and then down into proper alphabetical order. And I can't imagine what a fun job that must have been. I have done my share of indexing back in the pre-computer days when family history societies were indexing census returns, and you'd go home from the society meeting with a box full of paper slips, and you'd sort them and take them back next meeting. That's the same principle. So once they were sorted, you would have great big bundles of slips in alphabetical order, and then they would go to the indexers. And the indexers were the clerks who wrote out the original parchment indexes. And from 1837 up to 1865, they were handwritten parchment index volumes. And because they were handwritten--and this is a slow process--there was only one copy of each. They did all have very lovely handwriting. I have to say that. They made mistakes, but probably no more than you or I would. And these volumes were still in daily use up until 2007 when the physical volumes were removed and everything was then on microfiche or online. So parchment is very tough stuff.

And then the index volumes would have been prepared with the information that we still see on the online indexes. And the volumes with the quarterly returns--they would be put in the vaults, and the index volumes would go to the search room. And I was very pleased to find an example of transcription slips. Now, this is just the top part of the page that had 10 items on it. This happens to be a birth one, and these are from about 1910 when that system changed, and they stopped using these.

And these are probably much more sophisticated than the original ones, but you can see they're preprinted with the volume number and the registration district, and they've got a rubber stamp for putting the numbers on. I haven't seen the very early ones, but I rather suspect they were a lot more labor intensive. It's worth noting here, too, that the entries are written in cursive script. You and I, if we were doing something like this, we would probably do it in block capitals, but this is cursive. So when you are looking at things and you think something may have been mistranscribed, even as late as 1910, you still have to think of letters that are confused in cursive, not letters that look similar when they're in block caps or printed. And also, you'll notice--I think this is rather clever--there is a gap between each entry so that, when they came with a great big paper guillotine to slice them up, if the writing had strayed a bit as it has here, you could still get a nice clean cut and not lose anything. So it's quite a clever system. I was quite impressed. Now, this came to an end in 1910. This is why they've got this specimen here, because they like to embrace new technology. And from 1910, they started using the modern marvel called carbon paper. This meant that the local registrars could actually use carbon paper. When they were making their quarterly returns, they would automatically produce the transcription slips at the same time. So that saves a lot of work, and it also saves a stage of copying, although anyone who has ever used carbon paper will know what happens to carbon paper when it's been used a few hundred times, so I'm not confident that the slips that were later being sorted were always of the highest quality.

Now, I mentioned, as per Mike Foster, that the marriages can be particularly problematic, and I have actual personal experience of this, and this is a case that I was working on a number of years ago. You can see from the issue date of this marriage certificate, which I got from the GRO--I ordered it in 1998. Now, I was looking for the marriage of Joseph Rhodes to Sarah Ridgeway. I had a birth certificate of a child, and in these days, before we had online indexes, you just had to look through the volumes and look for a man of the right name and a woman of the right name that were in the same quarter and had exactly the same reference number--that meant they were on the same page. And I looked and I looked, and I did all the right things. I went back and checked it again. And try as I might, I could not find that marriage. I could find John Rhodes and Sarah Ridgeway, so I ordered it just in case there was a mistake and the indexer had put John and meant Joseph. No, there's the certificate, and it does say John. Oh, blast. However, I did mention that there are such things as local indexes. They're a wonderful, wonderful project, which started up a number of years ago, in some counties. It's usually family history societies cooperating with local registrars, and so there are now a number of local indexes. And these are going to be better than the national indexes because they are indexing the original copies, which are in the local register offices. The very first county to do this was Cheshire, and it did take me a while before it occurred to me to maybe check the Cheshire BMD indexes just in case. I mean, how optimistic do you have to be? But I searched the Cheshire indexes, and lo and behold, I found the marriage of Joseph Rhodes to Sarah Ridgeway exactly this year. So I ordered a copy of the certificate from Stockport Register Office. And there it is. It's Joseph Rhodes--father's John. It's Joseph Rhodes and Sarah Ridgeway. Now, being a genealogist and being cynical and hard-bitten, I thought, well, yes, I'm sure this is right, but you can see this is handwritten because, on the whole, register offices don't have the technology to scan things onto certificate planks. So this is handwritten, and there was always a possibility that someone in the register office had made a mistake. But I know that they didn't because, when the Cheshire Parish Registers went online, some time after that, I now have a third copy of the same marriage. And there is the entry from the parish register for Stockport, and you can see it is Joseph Rhodes. So it was the copy that was sent to the GRO by the clergyman who performed the marriage, and I can recognize his handwriting now. I have refrained from hunting down his descendants and asking them to account for it, but this is exactly the sort of thing that can happen. And although the marriage indexes are less good, for the reasons I've explained, than the births and deaths, the good news is that, because so many of the marriages were in churches, you can only get at the GRO's copy through their system and the certified copies. But as with this one, many of the church copies are not only widely available in record offices or on microfilm, in the Family History Library--increasing numbers are online. So I would always encourage you, if you've got a church marriage, always go and look at the parish register copy if you possibly can because it might be different as in this case. Sometimes it might just be a slight difference. It might be a little bit of extra information, or you might just want to look at the signatures. And that is the site that you go to to find those local indexes. It's called ukbmd.org.uk. And there isn't a local index project for everywhere, but if there is one, I would strongly advise you to go and look at it because these are very, very good indeed. So Cheshire led the way, and a lot of other areas copied their precise model. So a big shout out for UKBMD even though it's not what I'm talking about at the moment, but they deserve it.

Now, we've already established that these were quite boring jobs--the monotony of constant writing during 9 or 10 hours every day on task work. Task work was--you were paid for the number of entries that you wrote, and that was obviously an incentive to work fast, which was also an incentive to maybe be careless, although their work was checked. And the clerks would only get paid if they did it right, and sometimes you can see, on the old manuscript vellum pages, where a senior clerk has been and checked, and they've written the number of errors that they found. The checking didn't get done as much as it should have done because they always had more work than they could cope with, but there was some checking done.

And so it was an extremely boring job and tends to stupefy and destroy the senses of those employed in such never-ceasing work. It was 9 or 10 hours a day, as we know. And they often worked 5 1/2 or sometimes even 6 days a week--so not a lot of fun. But I found this wonderful item in a magazine in 1899, which is the kind of leisure reading that I have, so--doesn't everyone? And this is a Somerset House clerk--once said that the tedium of his labor on the registry of births and deaths is often relieved by coming across a humorous juxtaposition of names.

Then the face of the clerk will be covered with a smile. Isn't he wonderful? And I think we all do that if we're looking at names all day, especially when we used to have to look at indexes. We'll have our own little private informal collections of amusing names--things like twins called Alpha and Omega and things which were not remotely funny at the time, but they are now like Minnie Cooper. One of my particular favorites--Philadelphia Cheesman. I like that one. So it was a boring job, but at least there was an occasional bit of light relief.

Now, I've reproduced this document here. This is a table that was in one of the documents that I used mostly for this. I have some reservations about this because that is the actual title--"errors made by five index compilers over six or seven years." It doesn't say exactly which years, and whether it was all five of them for all those years, and, yes, I did put all the numbers into a spreadsheet and add them, and they do come to this rather worryingly round figure of 20,000. So I do have some reservations about this. It's also undated, although I know roughly when it must have been. It was 1860s and '70s because I know who these men were. They were quite experienced clerks. I know exactly who they were because I have a database of all the clerks who worked for the General Register Office from 1837 up to 1886 and beyond. So Dave Annal was absolutely right what he said about me at the beginning. And there is what they call a more detailed analysis of errors. This shows you the kind of errors that occurred, and, although I still have reservations about it, it does make the point that they did do some checking, and they were aware of the sort of things that went wrong. Now, some of the errors were--they wouldn't be fatal errors--a slight misspelling of a Christian name or missing out a middle name wouldn't stop you finding something, although putting something in completely the wrong registration district--or, you know, they say union here--that might. But even allowing for that, considering the millions of entries that are being dealt with over the years, the proportion of errors is actually quite small. So though I'm rather dwelling on all the things that go terribly wrong, it's not actually all that bad, so please don't think badly of the clerks. Now, let's look at some actual indexes. This is a page from one of the original vellums. I photographed this while the original vellum volumes were still available, and this happens to be from a marriage index from 1858. And there was a mistake on there to start with before you even get to the names because the pages were preprinted with the column headings and the columns, and this is a marriage volume, but they accidentally used one that was meant for births so that the column headings have been crossed out. It didn't matter dreadfully because the information being collected was exactly the same no matter whether it was a birth, marriage, or death, and they didn't want to waste this stuff. The clerks who wrote these up--the indexes--not only did they only get paid when they completed something, and did it correctly, if they made such a mess that the parchment had to be discarded and thrown away, they could get fined for the value of the parchment because this was very expensive stuff. I don't know if this ever actually happened or if it was just a threat hanging over them. And certainly, I've seen lots of alterations and corrections on the vellum indexes because common sense will tell you it was probably much more important to get the job done as quickly as possible rather than to punish the clerks. And this is one from a bit further on in the same volume. You can see that there's a name that's out of order. And this is almost certainly the sorter's fault. You imagine the poor indexer when he's writing out the name, and he writes out this name, and then he turns over and goes to the next one and finds that it's out of order. So you sometimes find little arrows like this when you've got an entry that's out of place or sometimes a whole block of entries. And the other thing too at the very top of the original page, which is very hard to see--you could see it sort of in the flesh, as it were--but little pencil notes there--sometimes the initials of the senior clerk who had to check the work and maybe the number that he found. But if I looked really closely, I could probably figure out who this clerk was, but, even for me, that's probably going too far. Now, this was the form that the indexes were in up until 1865. 1866 onwards, they sent the sorted slips out to a printer in East London called Darling, and they typeset the indexes from the sorted slips. And you can see here that, in many ways, these are going to be easier to read because they're printed. And of course, they could now print multiple copies. With the handwritten ones, you can only do one at a time. But with printing, you could produce more. And they actually produced four so that they would have a backup copy when the other one wore out. Paper is a lot less durable than parchment, so the paper volumes--they've been replaced. They must have used up all the back room spare copies because, in latter years, you would often find that the volume--the bound volume you were looking at--was a photocopy of one of these. And every time they got photocopied, they got a bit fuzzier and a bit fuzzier. But they're still fairly legible, and they've now taken the fairly sensible step of just having the surname once instead of repeating it on every single line. So these are more pleasing to look at at a glance. You'll also notice, at the bottom, that you've got what's called a write-in. If an entry arrived a bit too late for some reason once the index had already been compiled, it could be added in. You can see little asterisks there where these entries are supposed to be and then the details at the bottom. And the one at the very bottom is--this is 1881 birth register. Lily Grey was born in 1881, but for various reasons, her parents neglected to register her birth at the time. But she did get it registered late in 1933. That's probably the latest registration I've seen. More often, it's somebody is about a couple of months late. But you know, all things can happen. So I rather like that one. In 1865--sorry, in 1866--the very first of these printed indexes reduced the names down to the first Christian name, and any other names were just initials. In 1867, they went back to two names, and then the rest were initials. The old manuscript indexes--all the names were there. But from 1867 onwards, you only get the first two names. And it's worth knowing this because, when you're searching online, if you put in somebody's full name, you won't get the result back if the index has only got initials in. So it's worth knowing the format of the indexes even if you're not ever going to actually look at the pages themselves. So that's a birth index from 1881. And this next one here--this is the next significant thing that happened. I've already told you about the indexing slips, and they were using carbon paper from 1910. The other thing they did in 1910 was bring the indexing--the printing--back in-house. So Darling's--the printers--they were compensated because their contract hadn't run out, and this was brought back in-house, and these indexes are now typed. And this happened on all kinds of indexes from 1910. In 1911, they started adding more information into some of the indexes. In fact, oh, what I forgot to mention when I was talking about the first printed indexes--so, from 1866, the death indexes now included the age at death, which was a big help. That was the only change to the indexes. The certificates didn't change at all. There were no changes at all to the information on any of the certificates until 1969. But in 1910, they started typing the indexes and went back down to one name and the rest in initials, which is quite annoying if you've got somebody with a common name, but a really good middle name.

But they also added in information--you get the mother's maiden name for births. And that's added from the March quarter--sorry, from the September quarter of 1911. And from March 1912, you get the surname of the spouse in the marriage index. So that does make searching and finding the right entry a lot easier from then on.

And that's how the indexes continued up until 1969 when a couple of things happened. One was that the birth and death certificates changed and contained more information. And this is a birth index from 1982, but this is exactly the format that they were in from 1969 up until 1983 when this manual process stopped. You can see there were some write-ins there. And in this period, a write-in is very often a reregistration. This is when--the law was changed in 1926 so that, from '27 onwards, if a child's parents weren't married when the child was born, they could reregister the child once they did marry as their own legitimate child. And you do find quite a lot of these additions. They're not always, but sometimes that's the explanation. But the information in the indexes didn't change. The format looks a bit different, but the information is the same. But the death indexes--they did change because the extra information that was in the death certificates--instead of the age at death, you got the date of birth of the deceased person. This is very, very good although--health warning--the information is only as good as the person registering the death. And even if it was a close family member, they did sometimes make mistakes. You'd be surprised just how often people would get the day and the month right because they knew when granny's birthday was, but they might get the year wrong. And sometimes they didn't even know that much. There is one entry on this page, which just says "about" because somebody can only guess at the age of the person who's died. I picked this page--this is from 1970, and I didn't pick this one at random. I picked it because it's actually got rather a famous death on it there at the bottom of that first column. That's Jimi Hendrix, who died in London in 1970. So that's his death entry. Now, we don't look at these indexes anymore. We look at online ones, and FreeBMD is the place that you should always go to start with. It's a wonderful site. It's crowdsourced. It's been going for a number of years, and it's very good for a number of reasons--one, it's free, which is always a good thing, but it's also extremely high quality. It only goes up to 1983 because, after that, the indexers were born digital, and their job was to transcribe from the paper parchment indexes, which are very often from the microfiche or microfilmed copies. The indexes were microfiched in I think about the 1960s--sorry, microfilmed in the 1960s, and they were quite good quality--microfiched in the 1980s, and they were of less good quality. Most of the FreeBMD transcribers were working from the fiche or film copies, which have subsequently been scanned and are online. So these things have been through a few stages of copying before they get to FreeBMD, but it's a wonderful site. It's an extremely powerful search engine, and you can do all sorts of wonderful things with it. And I'd strongly recommend that you just sit and play with this for a while. This is the sort of thing you can do. I'm very interested in the surname of Firminger. And this is--I've just got every Firminger birth in Maidstone, and that's just a screenshot of it. That's as far as you can go until you buy an actual certificate. But you can do some other things with it. If you click on the page number, you can see all the other entries on the page, and the little symbol with spectacles--that will take you to an image of the page that they were transcribing from. So if there's something that doesn't seem quite right, you can check it out for yourself. I did this again with the surname Dax, which is very rare--about one entry a year. And I've just given you a screenshot up to March quarter of 1864, but the next one is--what is going on? The Daxes have suddenly sprung to life. What is going on is quite simple. I didn't tell you, when I was talking about the manuscript indexes, some of them were retyped, so if you see--some indexes before 1866 were typed up, and they look much like the later ones. And here's a couple of pages these entries come from. So you can see a bit more closely--there's the entry Gilbert Elliot Dax--and then after that, Ada Caroline. These should be in alphabetical order. The penny may be dropping.

The typist should have typed another surname there. She should have typed--and they were all females at this point. She should have typed "Day." So Ada Caroline and all the other people below her--their surname is Day, but they're indexed as Dax. The top of the next page, it's all right again. From Ellen onwards, they're correctly indexed as Day.

Now, this is a very obvious one because we have a very, very rare name immediately followed by a very common name. But it does happen elsewhere, and it's just less obvious because the two names aren't one rare and one common. So something else to look out for. Finally, on the printed indexes--something which is not at all obvious when you are using online searching, and that is, at the end of every death index volume, you get unknowns. And these are the people who have died and are unidentified. If someone died a long way from home, and they didn't have their credit card and their driving license and all these things--and in the 1880s, they probably didn't--all they could do was register, you know, "man aged about 45." Occasionally, you get a bit of a name, and this could be some local figure that nobody knew their real name--somebody there called Policeman Jack. So if you can't find an ancestor's death, that's where it might be. Pick one. No, no, that's very bad genealogy, but it's a possible explanation. Now, if I were doing this talk a couple of years ago, I would've stopped here, but a couple of significant things have happened. The General Register Office--they're finally freed from this certified copies only. They have now put--they've they've done two things. They have done their own indexes from the quarterly returns, so their indexes are not derived from the existing indexes. They've gone right back to their originals. They've put the indexes on their own site, which is where you go to order certificates, and they've got births up to 1917 and deaths up to 1957--no marriages, just the births and deaths. The other thing they have done is that they started a trial, which was a temporary trial, and they've gone back to it, and it's continuing possibly indefinitely so that, instead of ordering a certified copy--I'll show you some search results. They appear at the bottom of the page. You can click on a link there to order a certificate, or you can click on the PDF, and they can email you a PDF copy. And that is about 2/3 of the cost of the certificate, and it's a lot quicker as well. A certificate costs 9 pounds, 25 sterling, and you should only ever order it through the General Register Office. If anybody wants to charge you more than 9 pounds, 25, don't go there. Order it from the GRO. And the PDF copies are 6 pounds. So that's a big improvement. There are some downsides. they've added the age at death to all the death entries, which is great, right back to 1837. Unfortunately, some of them--they realized that the number is not necessarily years. It could be weeks or months. But if you have--you can submit corrections, and there's no automatic way they can review all this, but that they are correcting when those people point it out. They've also added in the mother's maiden name on all of the births right back to 1837. Small downside is that the search engine is not particularly flexible. You have to put in a surname, and it's not brilliant with variants. It uses Soundex and something else a bit like it. So you can't always find things that you want. But there is some good news. If I were doing this a couple of months ago, I would've stopped here. But since then, Findmypast have started adding that mother's maiden name information to the early indexes. Now, it's not complete, so you won't find everything. But because Findmypast has a much more flexible search engine, you can do things without searching by surname. And just to conclude--back to my Firminger family that I'm very interested in--I was able to search on Findmypast for the births, before the mother's maiden name appears in the full indexes, I was able to do a search purely on mother's maiden name, and I found quite a number of entries. I already knew about some of these people. So that's the way that you can get some of the data that you would normally only find on the GRO's new indexes. You can find some of it on Findmypast, not because they've got better data, but because they've got a more flexible search engine. And I am very hopeful that things will continue to improve. So watch the GRO. Watch their site for news, and, obviously, watch family history press and bloggers and things because it's taken over a century and a half, but things are improving. So I hope that at least some of you will now go and maybe try something that you haven't tried before and hopefully solve some problems that you thought you couldn't. And thank you very much for listening. [APPLAUSE]

If anyone has any questions--oh, good, some of you have--they have set up a microphone. It's a little way back, but if you could use the microphone so that everybody can hear the question, and then I'll do my best to answer them. The lady in the front, if you've got a loud voice, we'll start with you while people are making their way to the microphone. I just wanted to say thank you for the PDF versions of the certificates. Well, don't thank me. Thank the General Register Office. But everybody is absolutely delighted. You know, we've been moaning and complaining about this for years. And in fact, I don't know anybody in the General Register Office, including at least one former registrar general, who thought that this system we've been stuck with was a good one. But it can take a while to get things to change. We are talking about the civil service, but things are moving, so yay! OK, somebody at the microphone got another question. Yes I actually have two questions if I may. One is did commonwealths follow similar practices as far as like a place like Bermuda--did they forward copies of registrations to England? That's a very good question. Yes, they did, but not necessarily. Anything that was a colony was in charge of its own registration, so most registrar entries wouldn't be sent back to England. Some would. If somebody who was overseas and actually usually in a foreign country rather than a colony or a commonwealth country, they might choose to get their entry registered back home in the UK. But they didn't have to. And anyway, how could you enforce it? So there are some overseas birth, marriage, and death indexes in GRO, but there are only a tiny fraction of all the Brits overseas events that took place. And that's a whole wonderfully complicated subject in itself. So the short answer is mostly probably not, but there might be some. What was your other question? What about before 1837? Before 1837, you are reliant on church registers of one kind to another--Church of England baptisms, marriages, and burials and various nonconformist churches. Church of England entries--they had a legal status the others didn't have. And actually, it was a lot of the nonconformist churches who were very keen on campaigning for civil registration so that their events could be legally registered and had the same status as the Church of England because, no matter what your denomination was, unless you were Jewish or Quaker, you had to marry in the Church of England, between 1754 and 1837, for it to be legal. So you can imagine how popular that was with the Catholics and the Baptists and so on. So there was no nice central registry. There's lots and lots of births and baptisms and marriages and evidences of this, but they're all over the place. We do--we do cover quite a lot of that in the book I mentioned at the beginning, available in all good bookshops and probably in some terrible ones as well. I don't mind if you get it out of a library. I'm not going to retire on the proceeds anyway. So it's a whole different area of research before 1837. And of course, that's still only England and Wales. Civil registration started in Ireland and Scotland much later. OK, next gentleman there. Yes, on FreeBMD's searching marriages, occasionally, when you click through to see the list of names, you'll get an odd number of names. Oh, yes. Do you have a strategy for finding that one that's not there? Yes. Yes, I do. Now, you should see two men and two women. However, because of the transcription process--and we've seen that things have gone through lots of people copying things by hand and things being photocopied and fiched and filmed--errors do creep in. So you will sometimes find that three people on a page--their entries are correctly transcribed with the correct volume and page number--but one of them might be illegible. In the search results on FreeBMD, you'll sometimes see something in brackets or square brackets or just a question mark because they can't make it out at all. And then sometimes you can just go and look at the index page and work out what it says, but then that only really works for the later period when you can see the spouse's name. For the earlier period, the first thing I would always do is look to see if you can find the church marriage because they're mostly church marriages, and if you can find that, that will be a better copy. The other thing that I sometimes do--and this is so nerdy--you can download the results from FreeBMD, so you can kind of reconstruct a whole quarter's results for a registration district. Don't laugh. You can--I've done this. And sometimes, if you download them, put them in a spreadsheet, and you can sort them into order. And then you can see where the obvious gaps are. So you can find your three that have got the same number, but there's a missing man or a missing woman, and then you might find that there's a woman all on her own, who looks like she might be the one. And sometimes you can tell by looking at the number that it's a sort of number that could easily be mistranscribed. And so I've quite often done that and then gone and looked at the stray person that looks like they might fit and seen that the image for their index entry--yes, it is a bit indistinct, but it really is three and not an eight and so forth. So yeah, that's one of the great things about FreeBMD is being able to download things, and you can sort of reconstruct them and sometimes find the missing piece of the puzzle. And then when you've done that, you can tell them. And then they can correct it if they look at it and think, yes, I agree with you that it's three and not an eight or whatever. And then so FreeBMD is constantly improving because they have a lot of transcribers, and they go back, and they add in all sorts of features. And people are always adding new corrections, so it's a wonderful, wonderful site. I was a bit skeptical about it when it started out. I thought, this will never work. And why don't you do something more useful instead? I have never been so glad to be wrong in all my life. It really is an absolutely marvelous tool to use, and so it should always be your first go-to. And I can't remember if I mentioned--because I've been rehearsing this, and I've already done this talk once this week--Ancestry take their birth, marriage, and death index data from FreeBMD up to 1915. After that, they've done their own indexing. Findmypast have done their own indexing right from the start. So as with any other record, if the same thing is in two or three different places transcribed by different sets of people, you will find some discrepancies. So the general rule, if you can't find it on one site, it might be worth trying another one. And I've done that. I've done little comparisons, and they're broadly the same. I harped on about all the things that can go wrong, but, overall, most people find most of what they want most of the time using the ordinary, old-fashioned just look it up, and find the records. But it's when you dig into it, and you get as--obsessed is such an ugly word--shall we say, focused, as I am that you can actually make sense of things because you understand how the records are created. And then you can work out what the problem might be, which, if you just parachuted in and saw an individual entry in isolation, you wouldn't be able to make any sense of it. Well, that's my excuse, and I'm sticking to it. OK, I can think that was--that looks like it was the last of the questions, so thank you. Thank you very much for listening and-- [APPLAUSE]

--and being an audience and giving some background noise because this was live streamed and recorded, so it'll be on YouTube to haunt me for the rest of my life. [MUSIC PLAYING]

Thank you very much.