My Wagner ancestors settled in the town of Polk, Washington county.   But at some point back about 1978, I learned that they previously owned property in the town of Richfield as early as 1846.   So I ordered the 1850 Washington county census microfilm on interlibrary loan and brought home a portable microfilm reader (on the school bus!) from my high school library.

I scanned page by page and found only one Wagner family in the town of Richfield in 1850.

Based on what I knew plus the 1860 census (ten years after 1850), here is what I thought I would find...

Nikolaus Wagner, age about 43, Germany Angela, his wife, age about 43, Germany Anna, age about 17, Germany Peter, age about 14, Germany John, age about 11, Germany Nickolas, age about 8, Germany Martin, age about 4, New York Elizabeth, age about 1, Wisconsin

Instead, I found...

N. Wagoner, male, age 44, Germany M., female, age 42, Germany Dorathy, age 17, Germany Samuel, age 13, Germany George, age 11, Germany John, age 9, Germany Nicholas, age 5, New York Mary, age 2, Wisconsin

It sure looked like them, but the names just didn't jive.  So I talked myself into thinking that either the census taker missed 'my' Wagner family or they stayed in New York and just owned the land, sold it to Friedrich Becker and bought the other farm in the town of Polk before moving to Wisconsin (which seems silly now).

Spoiler alert... I hadn't noticed at the time how the genders, ages and birthplaces (see above) were almost an exact match for my family!

Then about the year 2000, I decided to try again. But this time I took the 1859 plat map and made note of where the census taker was on the map at each household.

That's when things got interesting!

Starting four households prior to that Wagner family and going four households beyond, the heads of the households were:

C. Melius ... N. Smith ... M. Peters ... P. Bong ... N. Wagner ... P. Roser ... J. Lawfer ... P. Shultis

Below are sections 3 thru 6 of the 1859 Richfield plat map (nine years after the 1850 census). The X marks the spot of the land owned by the Wagner family from 1846 to 1854. They sold it (prior to this map) to Friedrich Becker.

Two things came out of this.  First, I convinced myself that it WAS my family.   Second, when I shared my discovery with my friend Mary (a descendant of the Lofy family), we realized that she had not yet found her family in the 1850 census due to the surname misspelling... Lawfer verses Lofy, not to mention the first names being wrong. Side note - C. Hoelz/Helz above was the son-in-law of P. Rosar/Roser.

Other genealogists pointed out that sometimes children were referred to by their middle name.   But I had since found the Wagner baptisms in Germany and none of the children were baptized with middle names, particularly not 'Dorothy', 'Samuel', etc...

About a year later, I did some research for a lady whose ancestors lived near Hubertus (a little further south within the town of Richfield).   She wrote 'John, I think I found my family in the 1850 census, but the first names of the children are all wrong. What do you make of this?'

Having since researched in German records for many of the families in Richfield, I came to the realization that the ENTIRE township is this way.

Other genealogists pointed out that often the census taker couldn't understand German farmers, but it turns out all of the Irish families in the southern third of the town have the same problem!   The farmer's name is generally correct; the wife is listed as just 'Mrs.' or is incorrect and ALL of the children's names in the whole township are wrong.

My suspicion is that the census taker was lazy.   When visiting each farm, he only wrote down the ages, genders and birthplaces of the wife and the children.   When he went to turn his work and get paid, they told him he needed first names, so he took the papers back home and went line by line making up names for the wives and children based on each person's gender.

I recently came to the realization that the same person also took the census in 1850 for the town of Erin.   And it seems that before he finished making up first names in Erin, he got tired of doing so, and thus the last few pages of Erin in 1850 look like this...

It turns out that for three months that summer/fall, he enumerated four townships... Erin (from July 6 to Aug. 6), Richfield (from Aug. 7 to Aug. 19), Hartford (from Aug. 19 to Aug. 30) and Polk (from Sep. 2 to Sep. 30).   Yet Hartford and Polk have (somewhat) correct first names in them.   So I'm thinking it was after he completed Richfield and before he started Hartford that he learned that first names were required, so he did a better job when he set out in Hartford and Polk.

Important for genealogists...  If you have put additional children into your family (if they lived in Erin or Richfield in 1850) and assumed that these children died between 1850 and 1860 (or just disappeared), this census taker is the reason you have been lead astray.

The census taker was J. S. Loomis. He lived in Port Washington and in the page of the census on which he and his wife are enumerated, the other households are also from Port Washington, but the top of the page is incorrectly labeled as the town of Polk.   Vice versa, there are multiple pages of families who lived in the town of Polk, but the top of the page is labeled 'Port Washington'.