Family Tree
Research Log
Discovery
Research Plan
Field Guide
§1 The 1870 brick wall
The 1870 US Census is the first federal census that lists formerly enslaved Black Americans by their own full names. Before that, enslaved people were almost never named individually in official records — the 1850 and 1860 Slave Schedules list only the enslaver's name, with the people they enslaved reduced to tally marks: age, sex, and color. This is the single biggest reason Black family history research stalls before 1870, and it's not a gap in the records you're using — it's a gap in what was recorded at all.
§2 Freedmen's Bureau records (1865–1872)
The Bureau of Refugees, Freedmen, and Abandoned Lands operated field offices across the South from 1865–1872, and its records are the richest source for the immediate post-emancipation years: labor contracts, marriage registers, hospital and ration rolls, complaint registers, and letters from freedpeople themselves searching for family members sold away from them. Records are organized by state and then by field office, not by name — so you generally need to know the county or nearest office before you can search effectively. That's what the Discovery tab is for.
§3 Freedman's Savings & Trust Bank records (1865–1874)
Separate from the Bureau, the Freedman's Bank opened branches in most Southern cities and kept detailed depositor registers — often including birthplace, plantation, former enslaver's name, spouse, children, and siblings. When an ancestor's account survives, it can be the single most valuable document you find, because it's one of the few sources where a formerly enslaved person's own account of their family was recorded in their own words.
§4 Cohabitation records (North Carolina and other states, 1866–1868)
After emancipation, most Southern states required formerly enslaved couples who had been living as married to formally register that marriage with the county — since slave marriages had no legal standing before the war. In North Carolina, this followed the 1866 "Act Concerning Negroes and Persons of Color," and the resulting cohabitation bonds, filed with each county's Register of Deeds, list the couple's names, ages, and how long they'd been together — and sometimes the enslaver's name and number of children. About 20,000 of these survive across 52 NC counties.
§5 WPA Slave Narratives (1936–1938)
Federal Writers' Project interviewers recorded over 2,300 formerly enslaved people in the 1930s, capturing firsthand memories of specific plantations, enslavers, and family members. Coverage is uneven by county, so check neighboring counties too if your target county has none.
§6 Tracing through the enslaver's surname
Many formerly enslaved people kept the surname of a former enslaver, took the name of a different prominent local family, or chose an entirely new one at emancipation — so a shared surname is a lead, not proof. Once you have a candidate enslaver, their own estate records (wills, probate inventories, tax lists, bills of sale) can list enslaved individuals by first name and sometimes age, which you can then try to match against your family's oral history and later census entries.
§7 U.S. Colored Troops service and pension files (1863–1930s)
Roughly 180,000 Black men served in the U.S. Colored Troops during the Civil War, and their records are among the richest in this field. Start with the free NPS Soldiers & Sailors database to find a serviceman by name and unit; his compiled service record and — above all — his (or his widow's) pension file at the National Archives can run to dozens of pages of sworn testimony naming parents, spouses, children, former enslavers, and plantations. Widows' pension applications often required proving a slave marriage, so they preserve exactly the family detail no other record kept.
§8 DNA as a complementary tool
Because the paper trail often breaks at 1870 — and almost never names an African village — DNA matching (autosomal testing plus shared-match tools) is often the only way to narrow an African region. Ethnicity estimates point to areas (e.g. Nigeria, Ghana, Senegambia); matches with people who have deep paper trails in West Africa or the Caribbean can refine that further. Record company, estimates, and key matches on each person's DNA workspace. DNA works best alongside the paper trail, not instead of it — and a percentage is not a village.
§9 Bridging toward Africa
You cannot usually surname-search Africa the way you search the 1870 census. The workable path is: push the American or Caribbean line as far back as documents allow → capture any “African-born,” ethnonym (Igbo, Coromantee, Congo…), ship, or port clue → use DNA for a hypothesized region → search Slave Voyages by coast/decade/port and African Origins if an African given name survives. Tag every claim with confidence (documentary, DNA-supported, oral, or speculative). The Research Plan’s final step walks this sequence per ancestor.
§10 Example: starting points for Belmont / Gaston County, NC
To show how the U.S. side of the work fits together, here's what a real starting point looks like for one specific place — Belmont, in Gaston County, North Carolina.
Gaston County, NC — confirmed resources
- Gaston County's cohabitation records survive as part of its county marriage records — start with FamilySearch: North Carolina, County Marriages, 1762–1979.
- North Carolina's Freedmen's Bureau field office records (NARA microfilm M1909) are indexed and imaged at FamilySearch: NC Freedmen's Bureau Field Office Records. Charlotte (the nearest major field office to Gaston County) had a Freedmen's Hospital with its own patient registers.
- The NC State Archives' digitized Cohabitation Records collection covers the broader 1866–1868 statewide set, ongoing digitization.
- As of recent library research, no WPA narrators were recorded from Gaston or neighboring Lincoln County specifically — worth checking bordering Mecklenburg and Cleveland counties instead.
This same approach — find the county, find its surviving record types, then search each one by name — is what the Discovery tab automates for any state or county you enter.