By John

The recent effort of Ancestry to correct some of its birthplace entries in the census database is to be applauded. For many years we have known that census takers wrote "IA" for Indiana in the birth fields on the 1850 and 1860 censuses. Abstractors transcribed it incorrectly as "Iowa." A recent announcement herald that these earlier errors have been corrected.

There is still more to be done to make federal census images more complete. When Ancestry and other companies digitized the federal census records, they used the microfilm versions of those records, instead of going back to the original volumes. (The originals are only available for early census to 1870; the later schedules have been destroyed). Unfortunately, when the original films were created, some errors occurred and a few scattered pages here and there were never filmed. They still exist in the original volumes in the National Archives and Records Administration, but researchers using the census databases will not find the names on those pages, since they have never been digitized.

Here's a story that illustrates the challenge. More than a decade ago, the late George Fitzgerald, a local Fort Wayne genealogist, undertook some research on George McCulloch, a Fort Wayne banker and later Secretary of the Treasury in the Lincoln administration. He searched the 1850 census, but couldn't find him (a younger man with that name was listed, not the 40-year-old banker from Maine). Where was Hugh? George went to the National Archives in Washington, D.C., examined the original 1850 register for Allen County, and discovered that page 159 of Wayne Township, Allen County, had never been filed due to camera operator error. There he found Hugh McCulloch along with six other households, all of which were later transcribed and published in the "Allen County Lines," vol. 26, no 4 (June 2002).

In my research over the years I have discovered that other census pages were inadvertently missed. In 1820, a number of pages were left out of the microfilm of the Virginia census, now transcribed online, as well as in the "Virginia Genealogist," vol. 18, no. 2 (April-June 1974). Unfortunately, the accompanying household data was not abstracted.

I feel certain that there are other missing pages here and there in the microfilm editions that resulted from camera operator error many years ago. If a company such as Ancestry were to digitize these lost pages, they could truly lay claim to having the most complete census records extant. However, locating and correcting these omissions could be a herculean task.

In any case, it is worth noting that if you absolutely cannot find someone who you feel should be on the federal census, it may be that you are encountering a missing page. While not a widespread problem, such errors do exist.