How to interpret differences between different birth indices in England and Wales

Question

While exploring the possibilities of the new GRO indices, I've come across an anomaly that I can't (yet) explain.

For example:

Various databases based on the 'old' GRO indices show a total of 5 births of Mary Harper in Dudley in 1860. For example, Freebmd.org.uk (cross-checked against the images of the old indices) has:

Surname First name(s) District Vol Page

Births Mar 1860

HARPER Mary Dudley 6c 9

Births Dec 1860

HARPER Mary Ann Dudley 6c 15

HARPER Mary Ann Dudley 6c 114

HARPER Mary Ann Dudley 6c 16

HARPER Mary Ann Dudley 6c 6

WestMidlandsBMD which is supposedly based on the records held locally (which are the precursors to the GRO indices) has the following (no quarters indexed):

Surname Forename(s) Sub-District Registers at

HARPER Mary A Dudley (DUD) Dudley Register Office

HARPER Mary A Rowley Regis Sandwell Register Office

HARPER Mary A Rowley Regis Sandwell Register Office

HARPER Mary A Rowley Regis Sandwell Register Office

HARPER Mary J Dudley (DUD) Dudley Register Office

And the new GRO index of births has:

Name: Mother's Maiden Surname:

HARPER, MARY DAINTY

GRO Reference: 1860 M Quarter in DUDLEY Volume 06C Page 9

Having read Michael Whittingfield Foster's two books on "A Comedy of Errors: The Marriage Records of England 1837-1899" (which analysed the various stages in the record keeping process and assessed the opportunities for errors to creep in) I can well believe there were errors introduced in the process of transferring records from local to central records (and even before that in the process for transferring marriage records from a church to the local registrar).

However, in this case the local records seem to tie up with the old GRO indices (5 births) but the new index only has one.

Was the local index flawed, and the error transferred to the central index? Have entries been stolen from the central index (I have heard of pages being ripped out of the central registers)? Is there any other explanation that I've overlooked?

Answer 1

I have observed a similar problem with the Massachusetts State Archives' online searchable database for Vital Records, which can be accessed here: Search Vital Records (1841 - 1910). ("Vital Records" is the US term for birth, marriage, and death records.) The article Vital Records (1841-1910) Contents describes the scope of the online searchable database, but it doesn't discuss how the searchable index was produced.

Just like the situation we have with the GRO indices, it isn't always easy to find out whether the online index was made from the printed indexes which already existed or by going back to the original registers and making a fresh index.

Print indexes can found for some date ranges via Ancestry and on AmericanAncestors.org, the website of the New England Historical and Genealogical Society (NEHGS). NEHGS is the source of some of Ancestry's collections, as you can see from this source information for Massachusetts, Marriage Records, 1840-1915 (dbid=2511).

Ancestry.com. Massachusetts, Marriage Records, 1840-1915 [database on-line]. Provo, UT, USA: Ancestry.com Operations, Inc., 2013. Original data: Massachusetts Vital Records, 1840–1911. New England Historic Genealogical Society, Boston, Massachusetts. Massachusetts Vital Records, 1911–1915. New England Historic Genealogical Society, Boston, Massachusetts.

There is also a small overlap with Massachusetts, Marriage Index, 1901-1955 and 1966-1970 (dbid=2966)

Ancestry.com. Massachusetts, Marriage Index, 1901-1955 and 1966-1970 [database on-line]. Provo, UT, USA: Ancestry.com Operations, Inc., 2013. Original data: Department of Public Health, Registry of Vital Records and Statistics. Massachusetts Vital Records Index to Marriages [1916–1970]. Volumes 76–166, 192– 207. Facsimile edition. Boston, MA: New England Historic Genealogical Society, Boston, Massachusetts.

Comparing all these disparate sources with each other, and with the registers themselves on Ancestry and on FamilySearch can yield very different results.

For example: I have a clipping from the newspaper for the marriage of Charles Richard Herre and Emma Rose Bahler (whose surname was spelled as Poehler later in life) in Holyoke in 1895. My first searches of the online index at the Massachusetts state archives for their marriage gave me zero results. Searching for all women named Emma married in Holyoke in 1895, I found the bride's entry:

Last name First name  Location    Year    Volume  Page    Type
Bahler    Emma Rose   Holyoke 1895    451 667 Marriage

An equivalent search for anyone named Charles or Richard married in 1895 in Holyoke yields no results. Widening the search for all 1895 marriages in Holyoke, and looking at the results in surname alphabetical order shows this in the results:

Last name First name  Location    Year    Volume  Page    Type
Frebert   Carl George Holyoke 1895    451 652 Marriage
Kennedy   Bridget Holyoke 1895    451 657 Marriage

In other words, there are no entries for Holyoke in 1895 which have surnames in between Frebert and Kennedy -- part of F and K, and the entirety of the surnames for G, H, I and J are missing.

Unfortunately, with the way the GRO's searchable online index is set up, the surname field is mandatory, so you can't use the same technique of searching for first names only to reveal surname gaps in the database. But it might be possible to widen the search in other ways to reveal coverage gaps for particular geographic areas.