Alice Bell Snider (b. 28 Jul 1867, d. 8 Feb 1945 in Henry Co., Iowa) is listed in family records (and indeed in other people's records online) as the daughter of George W. Snider (1839-1905) and Martha Davison (1845-1908). The only problem is that George and Martha were married in Nov 1874, and waiting seven years after the birth of a child seems incredibly unlikely in conservative rural Iowa.

In fact, the 1870 Census shows George with a 3-year-old Alice and infant brother Edward (which matches up), but wife Mary Snider, age 26. By 1880, Alice is 12, Edward is 10, there is a new 1-year-old Orville — and wife Martha, age 34.

This all seems completely consistent with second marriage — and I found a 1865 marriage record that matches: George W. Snyder and Mary Greer, Washington Co., Iowa, United States . The change from i to y is a bit suspect, though, for official records (both census records are Snider). The 1870 census has the family in Missouri (confirmed by the 1885 Iowa census which has that as Edward's birthplace). So, perhaps they moved to Missouri after Alice's birth and then the family returned after Mary died?

Besides that, though, it's a mystery. Other than that marriage record and the census clues, I can't find any mention of her anywhere. Where might I look next in a case like this, where the first wife is apparently "masked out" by the second?

  • I have no idea where to look, coming from the wrong side of the Atlantic, but in a case like this, I would regard it as important to find proof of Mary's death as a first step in checking consistency. Otherwise, with the moves, there is still a faint chance there is only one wife (inconsistently named) and one marriage.
    – AdrianB38
    Commented Dec 17, 2013 at 22:59
  • @LearnedMan Sorry, I don't follow. What late birth?
    – mattdm
    Commented Apr 17, 2020 at 20:03

3 Answers 3


I agree that a second marriage is the most likely explanation. Snyder/Snider is not much of a difference.

Do you have any further records for the elder children, Alice and Edward, such as their marriages or deaths? That may show the name of their mother. Maybe they will mention their possible stepmother (Martha) rather than Mary Green/Greer, but given their ages when he remarried they might not.

Are there any death records in that area at that time? If you find a death of a Mary Snider/Snyder in 1870-1878 that may have more details.

"Other peoples records online" is a very suspect source, I would generally ignore that except for clues that can be verified elsewhere. The number of trees with the same information makes no difference, they are often copied from each other, rarely with any checking.

  • As Rob says, the children's marriage records or death certificates are probably the best first place to look, albeit not certain. A somewhat longer shot - if you've identified Mary Greer's parents, it's possible that her father's will might mention his deceased daughter's children by name. Wills are chancy - some are brief (provisions for the widow, and the rest to the children "share and share alike", without naming them), a few are more detailed (listing daughters' husband's names, names of deceased children's grandchildren, etc.). It's a bit of a crap shoot, but sometimes pays off.
    – cleaverkin
    Commented Dec 24, 2013 at 19:20

Census records are highly useful, as we well know.

However, the data is only as good as its source. It's not only the enumerators that got things scrambled, but the informants as well. I have no source for this, but I suspect that most errors can be laid at the feet of informants, though they did not mean to leave us befuddled.

Even when the informant was the head of a family or the spouse, vital record dates were just not very important to most people in the 19th Century. They may not have wanted to dig into the trunk in the attic to get the correct dates while the enumerator waited impatiently in the doorway. Errors would have increased when the informant was an in-law or a household child or even a neighbor from the farm down the road.

Transcription errors are a whole other can of wax.

In summary, don't marry yourself to census data when it could be very wrong. Is it possible that you're experiencing a degree of informant failure?


(Some of this answer was previously posted as an answer to the question How to find wife's maiden name in New England in the 1700s?, but I have added material which is specific to Iowa.)

The Family Search Wiki has several reference articles on researching women ancestors.

Information in this Wiki page is excerpted from the online course Research: Grandmothers, Mothers and Daughters-Tracing Women offered by The National Institute for Genealogical Studies. To learn more about this course or other courses available from the Institute, see our website. We can be contacted at [email protected]

Of the references listed there, I am most familiar with this book:

Schaefer, Christina Kassabian, The Hidden Half of the Family: A Sourcebook for Women’s Genealogy (Baltimore, Maryland: Genealogical Publishing Co., 1999).

For each of the 50 states, the author provides a timeline for important dates in state history, changes in the laws governing marriage and divorce, property and inheritance. She suggests where to look for records, offers a bibliography, and lists resources for studying women's history. The introductory material, including "how to use this book", discusses the problem of searching for records pertaining to women. Iowa is listed on page 116-119. One online resource listed there is the Iowa Women's Archives at the University of Iowa Library.

Most of the time, I learn maiden names of women by finding records which list people in family groups -- anything which is likely to name a person along with the names of their daughters, their parents, or their sisters (both married and single). Probate records and other court documents can be especially useful, as well as newspapers. Schaefer's book is a useful finding aid for many of these records.

For instance, her section "where to find Marriage and Divorce Records" gives us the timeline of record availability. County clerks began recording marriages in 1834. Statewide registration begins in 1880. So it is not surprising to see that the record you've been able to retrieve from FamilySearch was recorded at the county level. Keep in mind that a prior marriage may not have been in the same county. If his marriage to Martha is a second marriage, might George's first marriage have been in a neighboring county?

Also, don't forget that counties, like people, are 'born'. Have you checked when the counties in this area came into being? It's possible that a person like George could be married twice in the same place, but if a new county was created in the interval, the first marriage could be recorded in one county, and the second marriage could be recorded in another. Have you looked for church records?

Schaffer says several territorial and state censuses are available at the Iowa State Historical Society, which has since been renamed the State Historical Society of Iowa; the 1838 territorial census is available on microfiche at the University of Northern Iowa Library.

Whenever I get stuck, I like to read case histories from other people looking in the same place and time period to get clues from their descriptions of how they solved their problems. Books and articles like this, and others where genealogists detail their own research, can be a good place to get ideas of how and where to search. Can you find bloggers whose work is in the same area in Iowa?

Studying friends, associates, neighbors can also be a good way to build a picture of the entire community and can sometimes yield records that you missed because of poor indexing. See Elizabeth Shown Mills' Quicklesson 11: Identity and the FAN principle.

  • The 1838 Territorial Census is not relevant to this question -- I included the link for others who may refer to this answer for information about Iowa.
    – Jan Murphy
    Commented Mar 12, 2014 at 17:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.