Silas Pruett shows up once. The 1850 census lists him in Troup County, Georgia, thirty-four years old, born in South Carolina, farming with a wife and three children. Then nothing. No 1860 entry, no marriage record, no death record, no headstone anyone has photographed. For most descendants this looks like the end of the line. It isn't. People this well documented once don't stop existing. They stop existing where you're looking.
Check whether the map moved before you assume he did.
Troup County gave up land to form Meriwether County in 1827 and Harris County in 1828. County lines across the American South and Midwest kept shifting through the entire nineteenth century: a new county carved from an old one, a courthouse relocated, a parish boundary redrawn after a diocese reorganized. If Silas Pruett's land sat near either border, the 1860 enumerator may have counted him in a county that didn't exist a decade earlier, filed under a jurisdiction that has nothing to do with where he actually lived. Before searching for a person who moved, check whether the county did. The same logic holds outside the United States: a Prussian Kreis, a French arrondissement, an Italian comune all redrew their lines more than once. Record sets are organized by jurisdiction, and jurisdiction is not fixed.
The name on the record is rarely the name you typed.
Historical spelling was not standardized, and neither was the clerk writing it down. Pruett shows up as Prewitt, Pruitt, Prewett, and Pruatt within the same three counties in the same decade, sometimes for the same man across two different documents. Add a transcriber reading faded handwriting a century and a half later, and a fifth variant appears in the index that was never in the original hand. Search the record, not the spelling: use wildcard searches, browse the actual page images county by county instead of trusting the index alone, and try the name phonetically (Soundex exists for exactly this reason). A dead end at "Pruett" is very often a live lead at "Prewitt."
Look at the people standing next to him.
When a direct search fails, search around the person instead of for him. Siblings, in-laws, godparents, and the witnesses who signed a marriage bond or a land deed tend to live near each other, marry into the same handful of families, and appear together across decades of records. If Silas Pruett had a sister who married into the Whatley family and stayed put, her household in 1860 may list a brother-in-law, a nephew, or a stray reference to "S. Pruett, deceased" in a probate file that never generated a record of its own. Genealogists call this the FAN approach: family, associates, and neighbors. It works because a person who disappears from the record rarely disappears from the community.
Follow the deed, not just the death record.
A death record was never guaranteed to exist for Silas Pruett. A deed almost always was. Georgia counties, like most of the South and Midwest, kept deed books (grantor and grantee indexes recording every parcel bought and sold) far more completely than they kept vital records, and land changed hands constantly whether or not anyone bothered to register a birth or a death. Pull the grantor index for Troup, Meriwether, and Harris counties and look for Silas Pruett's name on the selling side. If he sold land himself in, say, 1855, he was alive and competent to sign that year. If instead the deed reads "heirs of Silas Pruett" or "administrator of the estate of Silas Pruett," a death window turns up without ever locating a death certificate: he died sometime between his last appearance as a living seller and that estate sale. Tax digests do similar work on a tighter clock. Georgia counties assessed real and personal property annually, and a name that appears on the 1858 rolls and vanishes from 1859 usually means one of three things: a death, a move, or land sold to someone whose name now carries the tax burden instead. Line up five or six consecutive years of tax rolls side by side, and the gap where a name drops out is often narrower than anything a census, taken only once a decade, could ever give you.
The record he left with his congregation, not his county.
Not every record that survives comes from a government office. If Silas Pruett or his children belonged to a specific church, a fraternal lodge, or a trade with its own mutual-aid society, that institution likely kept its own books, and those books answer questions the courthouse never asked. Baptist and Methodist churches across the antebellum South kept membership rolls and letters of dismission recording when a member left to join a congregation in another county, sometimes the only paper trail for a family's move. Fraternal orders (Odd Fellows, Freemasons, and later the Woodmen of the World) collected dues, paid death benefits, and logged both in ledgers a lodge secretary kept for decades, occasionally the only document anywhere naming an exact death date for a man with no headstone and no probate file. Tradesmen who belonged to an early union or guild left dues records and apprenticeship rolls the same way. None of this shows up in a standard vital-records search, because it was never a vital record to begin with. It's a membership record, kept by an organization Silas Pruett chose to belong to, which means finding it starts with a different question: not where did he live, but who did he belong to.
Sometimes the record itself is the thing that's missing.
Not every dead end is a search problem. Genealogists who specialize in this call it burned counties research, and it's a real, named category of loss, not a euphemism: roughly a fifth of Ohio's counties, and well over a third of Georgia's and Alabama's, lost courthouse records outright, mostly to courthouse fires. The entire 1890 US federal census met a similar fate: a 1921 fire in the Commerce Department building destroyed nearly all of it, so no county-level census exists for that year at all, no matter how carefully you search. A parish that stopped keeping registers during a war, an occupation, or a period of civil unrest leaves the same kind of hole. When this is what happened, the fix isn't a better search. It's substitute records: tax rolls, court minutes, land grants, militia muster rolls, and Bible records that were never centralized enough to burn along with everything else.
Check whether he's hiding in a different kind of record altogether.
If a person never shows up in the vital records you'd expect, ask whether they lived a life that vital records didn't track. Poorhouse and almshouse admission ledgers, county home registers, and asylum intake books recorded people who had no property, no church membership, and no reason to appear in a deed or a will. A man who never married in a church might still have a widow's pension file if he served in a militia or a war, since pension applications often collected affidavits from neighbors describing decades of a life that no other document preserved. These aren't fallback sources. For someone who was poor, transient, or institutionalized, they may be the only record that was ever created.
The newspaper named the relationship nobody else recorded.
Small-town newspapers did something almost no official record did: they described people in relation to each other, casually, in print, often years before or after the event a government record would have captured. A sheriff's sale notice for unpaid taxes had to run for several consecutive weeks by law, and it named the delinquent owner outright, sometimes with "heirs of" attached the same way a deed would. Probate courts frequently required an estate notice to be published, naming the administrator and inviting creditors and heirs to come forward, which occasionally lists relationships the probate file itself never spells out. The richest vein, though, is the smallest print: the social column. A line reading "Mrs. John Smith of Meriwether County, sister of the late Silas Pruett, visited relatives here last week" tells you something no census or deed ever will: a woman's maiden family tie, recorded under her married name, in a sentence nobody wrote for a genealogist to find a century and a half later. Digitized collections like Chronicling America, and a growing number of state historical society archives, have made searching these papers by name feasible instead of a matter of scrolling microfilm reel by reel. A name that never surfaces in a vital record search is worth running through a newspaper archive on its own.
Sometimes the wall means the relationship is wrong, not the record.
Before widening the search any further, ask an uncomfortable question: is the relationship on the tree actually the one that happened? A child recorded as a "son" in one household is sometimes a grandson, taken in after a daughter died young or bore him outside a marriage the family didn't want written down, and raised under a grandfather's name the same as any other child in the house. That kind of quiet reassignment leaves exactly the symptoms that look like a brick wall: no birth record where one should be, an age gap between father and son that's oddly wide, a sibling who's suspiciously older or younger than the pattern the rest of the children suggest. Test the hypothesis directly instead of assuming geography is the problem. Compare the ages of the supposed parents against a biologically plausible range for the child's birth year. Look for a daughter in the household who disappears from later records right around the time the "grandson" appears. A DNA test, where one's available, settles it outright. Widening a search across three more counties for a relationship that never existed is a much longer road than checking the one you already have.
Widening the search like this is exactly the kind of work AncestorIQ's Deep Research is built to do, without you having to hold five record types and three counties in your head at once. It checks the neighboring jurisdictions, tries the name variants, and pulls in the siblings and witnesses standing around your ancestor. It also follows the deed and tax trail when a vital record never existed, and checks the newspaper archives for a name mentioned in someone else's story. Then it shows you the actual record behind each one, so you can judge it yourself. The goal was never to find a name that matches. It's to find the person, wherever the paper trail happened to put him.
