A Clue that Nathan Isn't the Father?

Albert G. Lewis, my husband's 2nd great-grandfather, died on June 16, 1924, in Franklin, Fayette County, Pennsylvania. His death certificate shows that his parents were Nathan & Elizabeth Lewis. In the 1860 census, Albert appears in Nathan's household as a 14-year-old. At first glance, it all seems straight-forward, so it's no wonder that several Ancestry family trees display Nathan as the father of Albert.

But after looking a little more closely, something just doesn't seem right to me, and I'm not convinced of this father-son relationship. For instance:

• Albert was born in 1846 but doesn't appear in Nathan's household in the 1850 census. The year 1846 may be off, but every other census and his death certificate back up this year. 
• In the 1860 census, Nathan is 73 and Elizabeth is 65. If Elizabeth was Albert's mother, she would have been 51 years old when she gave birth to Albert and 57 years old at the birth of another little boy, Louis, who also appears in this census. Not impossible, but... 
• Finally, there's a 12-year gap between Albert and the next oldest child in the 1860 census.

Looking at Nathan's older children hasn't provided any answers so far; I haven't been able to find the youngest boy Louis Lewis in other records yet, and a newspaper search came up empty. If a will exists for Nathan in 1875, it may state his relationship with Albert, but Fayette County wills don't seem to be online, so I'm not sure yet if one exists. Orphan court records are available, but a look at Lewis surnames in guardian appointments did not mention an Albert. Church records may help, so I need to determine the family's religious affiliation.

I don't know how this will turn out, but as I've said before: Questioning can actually be a good thing and makes us better researchers. If you have any suggestions, or if you're a Lewis relative and have additional information to show that Nathan is indeed Albert's father, I'd love to hear from you!

Related Post: Be a Genealogy Skeptic