Irish Civil Registration Indexes at familysearch.org
Posted: Tue May 26, 2009 2:05 pm
The forthcoming issue of Irish Roots magazine will have a short article on this topic by Steven Smyrl. I compiled some notes on my own experience with this website and sent them to Steven when he was writing his article. Paddy Casey has encouraged me to post my critique here as it is "laced with Clare/Limerick references". So here it is:
The quickest way to get at the "Ireland, Civil Registration Indexes 1845-1958" collection is via http://pilot.familysearch.org/recordsea ... ionDetails
or equivalently via
http://tinyurl.com/cqjrlw2
[The direct URL appears to have changed slightly as of late April 2010 to the above from
http://pilot.familysearch.org/recordsea ... searchable which is now giving unexpected results.
The tinyurl above is a slightly modified version for the new address; anyone who has bookmarked the old tinyurl may need to update the bookmark by adding a 2 to the end of the original URL http://tinyurl.com/cqjrlw ]
The interface to the collection requires a recent version of Adobe (Shockwave) Flash Player to be installed on the user's computer. I found it necessary to adopt the "delete-and-reinstall" strategy before I could search the collection successfully. In Mozilla Firefox, you can check your installed Plugins by selecting Tools, Options, Main, Manage Add-ons..., Plugins. In Microsoft Windows, you can delete old versions by selecting Start, Settings, Control Panel, Add or Remove Programs and looking for Adobe Flash Player in the list of Currently installed programs. The latest version can be downloaded from http://get.adobe.com/flashplayer/
For many years, the LDS Church resisted pressure to migrate its popular Personal Ancestral File genealogical database software from MS-DOS to Microsoft Windows, arguing that many of its users were older and less wealthy people, unwilling or unable to replace their computer hardware and operating system software with the frequency that Microsoft would like. The current need for the latest software represents a dramatic change of policy. For those using the Microsoft Windows 98 operating system, the last version of Adobe Flash Player available is Version 9; pilot.familysearch.org requires Version 10 of Adobe Flash Player, so is apparently not available at all to those using Microsoft Windows 98.
I was recently in a cemetery with a cousin looking for the grave of a long-dead relative. She had a web-enabled mobile phone, so I thought it might be a good idea to check the civil registration indexes for a matching record. I was disappointed to find that Flash is no more compatible with such handheld devices than with older PCs, so we had no direct access to the indexes while we were in the cemetery.
Furthermore, the Adobe Flash Player interface disables virtually all the keyboard shortcuts in Mozilla Firefox, making it very frustrating for those who do not like a point-and-click interface.
The ability to highlight text in results and the normal Ctrl-C keyboard shortcut to copy information into another application are both annoyingly disabled; however, each entry includes a clickable Copy to Clickboard link which is a reasonable substitute.
The huge advantage of the new search facility is that it replaces roughly 1,033 separate indexes with a single index (19 annual indexes to non-Catholic marriages for 1845-1863; 14 annual indexes each to births, marriages and deaths from 1864-1877; and 324 quarterly indexes each to births marriages and deaths from 1878 to 1958). [Or are there some twentieth-century years with annual rather than quarterly indexes?] The reduction of 1,032 in the number of indexes to be searched will be particularly helpful to those engaged in a comprehensive one-name-study of an unusual surname.
Indeed, the number of occurences in the entire database of many unusual surnames is far fewer than the number of indexes that one previously had to search. For example, there are
25 hits (Exact and close matches) for Huleatt
26 hits for Ranalow
141 hits for Trousdell
154 hits for Smyrl
222 hits for Revington
348 hits for Blackall
819 hits for Phibbs
1,076 hits for Kett
However, the search results will never admit to more than 4,700 results for any search, e.g. for the surname Waldron. It is not clear how the 4,700 results are selected, but in the Waldron example they appear to comprise 2,845 births of children named Waldron, 686 births to mothers named Waldron, 1,165 deaths, but no marriage at all. (There is an unexplained discrepancy of 4 in these totals.)
The above tinyurl presents a search form which will at first be somewhat misleading. Some of the fields requested in the search form do not exist in the underlying database; others are interpreted strangely.
For example a search for marriages in 1912 for Mary Kelly produces 716 hits. However, 194 of these are in 1909 or earlier! The search is obviously covering a number of years either side of 1912, but without going through all 716 hits one cannot tell how wide a window is being used.
Having already found the entry for my Mary Kelly's husband, I just wanted confirmation that I had the correct entry before spending four euro on a photocopy, and was not interested in the vast majority of my search results. I should be able to do this by entering volume and page numbers in the search form, but this is not possible.
The equivalent search for England and Wales at http://www.freebmd.org.uk/cgi/search.pl
allows the researcher to click on the page number to view a list of all other names registered on the same page; this simple addition would add enormously to the value of the new facility, particularly for those seeking the name of an unknown spouse.
-----
To use the database productively, it is best to fill in at most the first name, last name and/or (for births in the relevant period) mother's maiden name on the search form. If this results in less than 4,700 matches, then one can use the "Narrow by" links at the top of the screen to refine the search results:
Place
Date
Gender
Last Names
First Names
Role
Event Type
A very nice feature of the results is that they show the counties covered by the Superintendent Registrar's Districts (or Poor Law Unions, but confusingly called "Registration District"). Many of these cross county boundaries; for example, the Limerick Union covers parts of south-east Clare as well as Limerick City and its environs in County Limerick. Likewise, Tipperary Union covers parts of County Limerick and parts of County Tipperary; and Roscrea PLU covers parts of Counties Laois, Offaly and Tipperary.
1. Clicking on the Place link produces a list of the counties covered by the PLUs in the search results. The natural thing to do with this list is to select the county where the subject of the search was known or thought to have been born, married or died. Sometimes, however, it is better to click a neighbouring county! For example, if searching for someone living in south-east Clare with a surname which is common throughout county Clare but less common in county Limerick, selecting County Limerick may produce fewer results that selecting County Clare.
Searches would be made much easier if there was a Place list containing the PLUs as well as the list containing the counties. As both pieces of information are contained in the underlying database, it is strange that this facility is not provided.
The place list also often includes "Unspecified"; this seems to cover results where the surname being sought appears as mother's maiden name on a birth registration.
2. Clicking on the Date link produces a list of the decades associated with the results. For every result, the decades include the decade in which the event was registered. For deaths, the decades also include an estimated decade of birth.
The age-at-death in the indexes has been used to estimate a year-of-birth for each death entry. All genealogists will be aware that the age-at-death is notoriously unreliable; furthermore, the calculation assumes that the age is not the "Age last birthday" requested on the certificate, but the age at the birthday in the year of death. Thus, even if the underlying data was accurate, on average half of all death entries would result in a year of birth one year too late.
As the underlying database contains a precise year (and, in most cases, a precise quarter) of registration, it is hard to understand why results can be refined only by decade.
Even when an event was clearly registed in a later decade, the Date shown is the decade of registration.
For example, the well-known lexicographer Tomás de Bhaldraithe (1916-1996) had the name on his birth certificate changed from Thomas MacD[onagh] Waldron to Tomás Mac Donnchadha de Bhaldraithe in 1944 (the year after his marriage). This produced two database entries in the 1940s, but none in the 1910s. The 1940s entries are:
Thomas MacD 1916 Waldron (Limerick, third quarter, Volume 5, Page 309).
Thomas Mac D de Bhaldraithe (Limerick, third quarter, Volume and Page fields inexplicably blank).
However, the first of these entries does show up in the First Names field, where the list of initials in a Waldron search (but not a de Bhaldraithe search) includes the numeral 1!
3. Clicking on the Gender link allows one to choose Female (i.e., entries arising as mother's maiden name) or Unknown (i.e., all other entries).
4. Clicking on the Last Names link produces a list of initials, which sometimes helps to identify double-barrelled names, e.g. on a search for Waldron, the initials H and K denote Waldron-Hamilton and Waldron-King respectively.
In the same search, L and O denote the death record of Mary Lally ors Waldron; "ors" (short for "otherwise") frequently appears in the marriage indexes to denote the re-marriage of a widow, but it is quite unusual in the death indexes. Whether it merits indexing under O is debatable.
Finally, M and S appear in the Last Names list for the death of "Columba Waldron Sr M", presumably a nun, Sister Mary Columba Waldron.
5. Clicking on the First Names link again produces a list of initials; as already noted, these may include a numeral (1) to draw attention to a late registration.
6. Clicking on Role is yet another way of distinguish the Mothers whose maiden names appear in the database from the Principals.
7. Clicking on Event Type allows one to select Birth, Death, Marriage or Unspecified. Selecting Death or Marriage, as expected, selects the deaths or marriages. Selecting Unspecified again selects birth records where the mother's maiden name matches the last name entered. But selecting Birth selects boths births and deaths, again because an estimated birth year has been calculated for each death.
-----
Clicking "Refine search" on the results page allows one to search the Ireland, Civil Registration Indexes 1845-1958 again. Clicking "New search" on the results page moves one to a search of "All Indexed Collections".
The interface allows one to search by first name as well as by last name. This has a couple of potential uses.
First, if an unusual first name, especially an unusual female first name (e.g. Hermina or Marcella) runs in a family, it can help to track down missing female lines.
Second, we can now see when first names began to be popular in Ireland. For example, only one entry in the nineteenth century had the name Fiona; it occured 17 more times up to 1939; 36 times in the 1940s; 209 times in the 1950s; and even three times in the 1960s.
Since the source data ends in 1958, the last result is quite a puzzle. In fact, a search for marriages in 1968 produces 433 hits; a search for births in 1968 produces 2,047 hits; and a search for deaths in 1968 produces 402 hits. Whether the source genuinely includes some data from the 1960s or whether there has been a high error rate in data entry is not clear. A search for deaths in 1975 produces 8 hits; three of these are in 1971 with no quarter given, which suggests that the second digit of the year has been mistyped and that the year should read 1871, when the indexes had not yet been split into quarters.
The indexes do not appear to include all the Irish civil records which have been in the International Genealogical Index for many years. For example, there are only 6 hits for Peter Gallivan, none in County Kerry. But the IGI at http://www.familysearch.org/eng/Search/ ... 0304008465 includes a Peter Gallivan from Baatch No. C005951, born on 06 JUL 1876 in Kerry, Ireland
Parents:
Father: Timothy Gallivan
Mother: Margaret Gallivan Shea
This may be one of the estimated 10-15% of all registered events which are omitted from the indexes.
Volume numbers are not all correctly transcribed. For example (see http://www.clarelibrary.ie/eolas/coclar ... ampion.htm), Bertram Foster Kerrison married Noreen Lyons in 1936 Q4 in Limerick Union. The database entries are:
Name: Bertram F Kerrison
Registration district: Limerick
Record type: MARRIAGES
Registration date - quarter and year: Oct - Dec 1936
Film number: 101577
Volume: 5
Page: 177
Digital GS number: 4199369
Image number: 00431
Collection: Ireland, Civil Registration Indexes 1845-1958
Name: Noreen Lyons Registration district: Limerick
Record type: MARRIAGES
Registration date - quarter and year: Oct - Dec 1936
Film number: 101577
Volume: 3
Page: 177
Digital GS number: 4199369
Image number: 00432
Collection: Ireland, Civil Registration Indexes 1845-1958
Those who are familiar with volume numbering may be able to figure out whether this entry should be in Volume 3 or Volume 5.
The quickest way to get at the "Ireland, Civil Registration Indexes 1845-1958" collection is via http://pilot.familysearch.org/recordsea ... ionDetails
or equivalently via
http://tinyurl.com/cqjrlw2
[The direct URL appears to have changed slightly as of late April 2010 to the above from
http://pilot.familysearch.org/recordsea ... searchable which is now giving unexpected results.
The tinyurl above is a slightly modified version for the new address; anyone who has bookmarked the old tinyurl may need to update the bookmark by adding a 2 to the end of the original URL http://tinyurl.com/cqjrlw ]
The interface to the collection requires a recent version of Adobe (Shockwave) Flash Player to be installed on the user's computer. I found it necessary to adopt the "delete-and-reinstall" strategy before I could search the collection successfully. In Mozilla Firefox, you can check your installed Plugins by selecting Tools, Options, Main, Manage Add-ons..., Plugins. In Microsoft Windows, you can delete old versions by selecting Start, Settings, Control Panel, Add or Remove Programs and looking for Adobe Flash Player in the list of Currently installed programs. The latest version can be downloaded from http://get.adobe.com/flashplayer/
For many years, the LDS Church resisted pressure to migrate its popular Personal Ancestral File genealogical database software from MS-DOS to Microsoft Windows, arguing that many of its users were older and less wealthy people, unwilling or unable to replace their computer hardware and operating system software with the frequency that Microsoft would like. The current need for the latest software represents a dramatic change of policy. For those using the Microsoft Windows 98 operating system, the last version of Adobe Flash Player available is Version 9; pilot.familysearch.org requires Version 10 of Adobe Flash Player, so is apparently not available at all to those using Microsoft Windows 98.
I was recently in a cemetery with a cousin looking for the grave of a long-dead relative. She had a web-enabled mobile phone, so I thought it might be a good idea to check the civil registration indexes for a matching record. I was disappointed to find that Flash is no more compatible with such handheld devices than with older PCs, so we had no direct access to the indexes while we were in the cemetery.
Furthermore, the Adobe Flash Player interface disables virtually all the keyboard shortcuts in Mozilla Firefox, making it very frustrating for those who do not like a point-and-click interface.
The ability to highlight text in results and the normal Ctrl-C keyboard shortcut to copy information into another application are both annoyingly disabled; however, each entry includes a clickable Copy to Clickboard link which is a reasonable substitute.
The huge advantage of the new search facility is that it replaces roughly 1,033 separate indexes with a single index (19 annual indexes to non-Catholic marriages for 1845-1863; 14 annual indexes each to births, marriages and deaths from 1864-1877; and 324 quarterly indexes each to births marriages and deaths from 1878 to 1958). [Or are there some twentieth-century years with annual rather than quarterly indexes?] The reduction of 1,032 in the number of indexes to be searched will be particularly helpful to those engaged in a comprehensive one-name-study of an unusual surname.
Indeed, the number of occurences in the entire database of many unusual surnames is far fewer than the number of indexes that one previously had to search. For example, there are
25 hits (Exact and close matches) for Huleatt
26 hits for Ranalow
141 hits for Trousdell
154 hits for Smyrl
222 hits for Revington
348 hits for Blackall
819 hits for Phibbs
1,076 hits for Kett
However, the search results will never admit to more than 4,700 results for any search, e.g. for the surname Waldron. It is not clear how the 4,700 results are selected, but in the Waldron example they appear to comprise 2,845 births of children named Waldron, 686 births to mothers named Waldron, 1,165 deaths, but no marriage at all. (There is an unexplained discrepancy of 4 in these totals.)
The above tinyurl presents a search form which will at first be somewhat misleading. Some of the fields requested in the search form do not exist in the underlying database; others are interpreted strangely.
For example a search for marriages in 1912 for Mary Kelly produces 716 hits. However, 194 of these are in 1909 or earlier! The search is obviously covering a number of years either side of 1912, but without going through all 716 hits one cannot tell how wide a window is being used.
Having already found the entry for my Mary Kelly's husband, I just wanted confirmation that I had the correct entry before spending four euro on a photocopy, and was not interested in the vast majority of my search results. I should be able to do this by entering volume and page numbers in the search form, but this is not possible.
The equivalent search for England and Wales at http://www.freebmd.org.uk/cgi/search.pl
allows the researcher to click on the page number to view a list of all other names registered on the same page; this simple addition would add enormously to the value of the new facility, particularly for those seeking the name of an unknown spouse.
-----
To use the database productively, it is best to fill in at most the first name, last name and/or (for births in the relevant period) mother's maiden name on the search form. If this results in less than 4,700 matches, then one can use the "Narrow by" links at the top of the screen to refine the search results:
Place
Date
Gender
Last Names
First Names
Role
Event Type
A very nice feature of the results is that they show the counties covered by the Superintendent Registrar's Districts (or Poor Law Unions, but confusingly called "Registration District"). Many of these cross county boundaries; for example, the Limerick Union covers parts of south-east Clare as well as Limerick City and its environs in County Limerick. Likewise, Tipperary Union covers parts of County Limerick and parts of County Tipperary; and Roscrea PLU covers parts of Counties Laois, Offaly and Tipperary.
1. Clicking on the Place link produces a list of the counties covered by the PLUs in the search results. The natural thing to do with this list is to select the county where the subject of the search was known or thought to have been born, married or died. Sometimes, however, it is better to click a neighbouring county! For example, if searching for someone living in south-east Clare with a surname which is common throughout county Clare but less common in county Limerick, selecting County Limerick may produce fewer results that selecting County Clare.
Searches would be made much easier if there was a Place list containing the PLUs as well as the list containing the counties. As both pieces of information are contained in the underlying database, it is strange that this facility is not provided.
The place list also often includes "Unspecified"; this seems to cover results where the surname being sought appears as mother's maiden name on a birth registration.
2. Clicking on the Date link produces a list of the decades associated with the results. For every result, the decades include the decade in which the event was registered. For deaths, the decades also include an estimated decade of birth.
The age-at-death in the indexes has been used to estimate a year-of-birth for each death entry. All genealogists will be aware that the age-at-death is notoriously unreliable; furthermore, the calculation assumes that the age is not the "Age last birthday" requested on the certificate, but the age at the birthday in the year of death. Thus, even if the underlying data was accurate, on average half of all death entries would result in a year of birth one year too late.
As the underlying database contains a precise year (and, in most cases, a precise quarter) of registration, it is hard to understand why results can be refined only by decade.
Even when an event was clearly registed in a later decade, the Date shown is the decade of registration.
For example, the well-known lexicographer Tomás de Bhaldraithe (1916-1996) had the name on his birth certificate changed from Thomas MacD[onagh] Waldron to Tomás Mac Donnchadha de Bhaldraithe in 1944 (the year after his marriage). This produced two database entries in the 1940s, but none in the 1910s. The 1940s entries are:
Thomas MacD 1916 Waldron (Limerick, third quarter, Volume 5, Page 309).
Thomas Mac D de Bhaldraithe (Limerick, third quarter, Volume and Page fields inexplicably blank).
However, the first of these entries does show up in the First Names field, where the list of initials in a Waldron search (but not a de Bhaldraithe search) includes the numeral 1!
3. Clicking on the Gender link allows one to choose Female (i.e., entries arising as mother's maiden name) or Unknown (i.e., all other entries).
4. Clicking on the Last Names link produces a list of initials, which sometimes helps to identify double-barrelled names, e.g. on a search for Waldron, the initials H and K denote Waldron-Hamilton and Waldron-King respectively.
In the same search, L and O denote the death record of Mary Lally ors Waldron; "ors" (short for "otherwise") frequently appears in the marriage indexes to denote the re-marriage of a widow, but it is quite unusual in the death indexes. Whether it merits indexing under O is debatable.
Finally, M and S appear in the Last Names list for the death of "Columba Waldron Sr M", presumably a nun, Sister Mary Columba Waldron.
5. Clicking on the First Names link again produces a list of initials; as already noted, these may include a numeral (1) to draw attention to a late registration.
6. Clicking on Role is yet another way of distinguish the Mothers whose maiden names appear in the database from the Principals.
7. Clicking on Event Type allows one to select Birth, Death, Marriage or Unspecified. Selecting Death or Marriage, as expected, selects the deaths or marriages. Selecting Unspecified again selects birth records where the mother's maiden name matches the last name entered. But selecting Birth selects boths births and deaths, again because an estimated birth year has been calculated for each death.
-----
Clicking "Refine search" on the results page allows one to search the Ireland, Civil Registration Indexes 1845-1958 again. Clicking "New search" on the results page moves one to a search of "All Indexed Collections".
The interface allows one to search by first name as well as by last name. This has a couple of potential uses.
First, if an unusual first name, especially an unusual female first name (e.g. Hermina or Marcella) runs in a family, it can help to track down missing female lines.
Second, we can now see when first names began to be popular in Ireland. For example, only one entry in the nineteenth century had the name Fiona; it occured 17 more times up to 1939; 36 times in the 1940s; 209 times in the 1950s; and even three times in the 1960s.
Since the source data ends in 1958, the last result is quite a puzzle. In fact, a search for marriages in 1968 produces 433 hits; a search for births in 1968 produces 2,047 hits; and a search for deaths in 1968 produces 402 hits. Whether the source genuinely includes some data from the 1960s or whether there has been a high error rate in data entry is not clear. A search for deaths in 1975 produces 8 hits; three of these are in 1971 with no quarter given, which suggests that the second digit of the year has been mistyped and that the year should read 1871, when the indexes had not yet been split into quarters.
The indexes do not appear to include all the Irish civil records which have been in the International Genealogical Index for many years. For example, there are only 6 hits for Peter Gallivan, none in County Kerry. But the IGI at http://www.familysearch.org/eng/Search/ ... 0304008465 includes a Peter Gallivan from Baatch No. C005951, born on 06 JUL 1876 in Kerry, Ireland
Parents:
Father: Timothy Gallivan
Mother: Margaret Gallivan Shea
This may be one of the estimated 10-15% of all registered events which are omitted from the indexes.
Volume numbers are not all correctly transcribed. For example (see http://www.clarelibrary.ie/eolas/coclar ... ampion.htm), Bertram Foster Kerrison married Noreen Lyons in 1936 Q4 in Limerick Union. The database entries are:
Name: Bertram F Kerrison
Registration district: Limerick
Record type: MARRIAGES
Registration date - quarter and year: Oct - Dec 1936
Film number: 101577
Volume: 5
Page: 177
Digital GS number: 4199369
Image number: 00431
Collection: Ireland, Civil Registration Indexes 1845-1958
Name: Noreen Lyons Registration district: Limerick
Record type: MARRIAGES
Registration date - quarter and year: Oct - Dec 1936
Film number: 101577
Volume: 3
Page: 177
Digital GS number: 4199369
Image number: 00432
Collection: Ireland, Civil Registration Indexes 1845-1958
Those who are familiar with volume numbering may be able to figure out whether this entry should be in Volume 3 or Volume 5.