His death certificate has his date of birth as 11 Feb. 1856 (the informant seems to be someone who worked at the hospital where he died--it's no relative that I've found): http://seekingmichigan.contentdm.oclc.org/cdm/ref/collection/p129401coll7/id/239262
Every census record of him in the US would have him born a decade earlier in the late 1840s.
1880: https://familysearch.org/pal:/MM9.1.1/MW3T-LVN
1900: https://familysearch.org/pal:/MM9.1.1/M9BY-LYW
1910: https://familysearch.org/pal:/MM9.1.1/MLTJ-8T9
His Find A Grave entry has the 1856 year and a copy of the death certificate posted but does not have a picture of the grave marker. A lot of the entries for this cemetery are based on the burial date on the cemetery's website cross referenced with the death certificate without necessarily viewing the grave marker or other sources: https://www.findagrave.com/memorial/163019848
I have not found a birth record or a marriage record for him yet either, but he married his wife in about 1878/79 (see 1900 census), which would make him either in his early twenties or early thirties, so either is plausible.
Right now I have his DOB as 11 Feb. 1846 and marked "uncertain." Should I keep it that way and add a note about the discrepancy in years from the census and the death record/Find A Grave? (Though now that I'm relooking, I'm thinking the year should be about 1848 uncertain based on the census records.) Or should I change the year to 1856 uncertain (and still add the note)?
I'm personally inclined to go with the dates on the earlier (census) records because I've found a lot of birthdays on death records, especially the early ones, that are inaccurate/impossible (e.g., child was supposedly born a month after her older sibling). With such a large discrepancy in years and no childhood sources, though, I thought it might be better to see what others thought about it.
