Irish Famine Begins 1845: Using DNA Matches to Find the Townland Records Never Recorded

Your immigrant ancestor’s death certificate says “Ireland.”

That’s the whole entry. No county. Nothing that would let you order a baptismal record or walk a graveyard. And every September, the anniversary posts roll around and remind you that the thing which drove her out of that unnamed place started on a specific day in 1845.

Except it didn’t. Not exactly.

Click HERE to get your copy of From Spit to Stories:
Using AI to Unlock Your DNA Ancestry by Thomas MacEntee!

Irish Famine Begins: The Date on the Calendar and the Date in the Fields

September 9, 1845 is the date you will see attached to the arrival of potato blight in Ireland. It shows up in reference works and in a fair number of genealogy blog posts every autumn. It is generally traced to a report in the Dublin Evening Post, and as a peg to hang a remembrance on, it works fine.

As history, it flattens something slower and considerably messier.

David Moore, curator of the Royal Dublin Society’s Botanic Gardens at Glasnevin, noticed the first signs of the disease on his potatoes on August 20, 1845. That is roughly three weeks before the date on the calendar. Contemporary accounts also place a total crop failure in County Fermanagh about a week after the Glasnevin sighting. Meanwhile the English horticultural press had been sounding alarms since mid-August, after a subscriber on the Isle of Wight described a blight of unusual character running through the island’s potatoes. Irish newspapers reprinted those English reports.

Then, on September 13, the Gardeners’ Chronicle stopped its press to announce that the potato murrain had declared itself in Ireland beyond any doubt. Four days after the anniversary date. A month after the first English warnings.

So which day did the Famine begin? None of them, probably. A crop disease arrives the way weather arrives, in patches, over weeks, noticed by one gardener here and one tenant there before anybody understands what they are looking at. Even in 1845 it was hard to assess the damage, precisely because the blight showed up across a spread of months rather than all at once. Treat September 9 as an anniversary that people found useful. It was never a documented starting gun. It is worth saying plainly, because Irish research punishes you for accepting tidy stories.

Irish Famine Begins: What It Cost, and Why the Numbers Are Argued Over

Roughly a million people in Ireland died of starvation and related causes, and at least another million left the country before the Famine ended in 1852. Ireland’s population dropped from nearly 8.4 million in 1844 to 6.6 million by 1851.

You will also see the figure of two million dead in commemorative material. That higher number is contested and tends to blend deaths with emigration, or to stretch the period. The million-dead estimate is the one most historians work from, with the honest caveat that nobody counted properly at the time.

Here is the part that lands on your research desk. The people who left were, in enormous numbers, cottiers and tenants at will. They held no deeds. They left no wills. By March 1851 nearly four percent of the population had been driven into workhouses, a system originally built to hold about one percent. Families like that generate very little paper on the way out, and what paper they do generate on the way in says things like “Ireland” and “Irish.”

Irish Famine Begins: Then 1922 Happened

On June 30, 1922, in the opening days of the Irish Civil War, two mines exploded in the Record Treasury of the Public Record Office of Ireland at the Four Courts in Dublin, and more than seven hundred years of records were destroyed in an afternoon. The census returns for 1821 through 1851 went up. The 1861 through 1891 returns had already been destroyed by government order before the fire, which means the earliest complete surviving Irish census is 1901. Somewhat over half the Church of Ireland registers deposited there burned, along with most wills and testamentary records proved in Ireland.

That is the gap. Your ancestor left in 1848, and the census that would have listed her household in 1841 no longer exists.

Do not overcorrect, though. The assumption that everything burned is wrong. Catholic and Presbyterian registers, among others, were never stored there. Fragments of the 1821 to 1851 censuses survive, and probate indexes still summarize the lost wills. Griffith’s Valuation survived, since it was printed and distributed rather than shelved in the Treasury. The Virtual Record Treasury of Ireland project has been reassembling substitutes and duplicates from archives on both sides of the Atlantic.

Irish Famine Begins: Why a Cluster Beats a Document You Do Not Have

DNA is indifferent to fire. Cousins descended from the people who stayed in the townland are walking around today with segments of the same chromosomes as cousins descended from the people who sailed. If enough of both groups have tested, the connection between them exists whether or not a single sheet of paper survived.

You already know what a shared match is. The move here is to stop treating shared matches as a way to confirm a relationship you already suspect and start treating a group of them as a geographic argument.

The logic runs like this. Take one cluster of matches who all share DNA with you and with each other. Some of them will have trees. A few of those trees will name an Irish parish, because their branch of the family kept a headstone or a marriage record with more detail than yours. If nine or ten people in a single cluster independently point at the same corner of the same county, you have a hypothesis about where your ancestor came from. It is a hypothesis and nothing more, though it is a testable one, which is more than “Ireland” ever gave you.

Irish Famine Begins: Working the Method

Start by pulling matches above 20 cM. That threshold is not arbitrary. Ancestry’s shared match list only displays matches at 20 cM and above, which is what they mean by fourth cousins and closer. Ancestry Pro Tools subscribers get Enhanced Shared Matches, which shows how much DNA your matches share with each other and makes smaller shared matches visible below that cutoff. For Famine-era work that extra layer earns its keep, since you are usually operating five or six generations back where relationships are ambiguous.

Group what you pull. The Leeds Method, developed by Dana Leeds in 2018, uses a spreadsheet to sort matches into color groups that often resolve into four clusters matching your four grandparent lines. Automated versions exist if spreadsheets make you tired.

Then isolate the Irish cluster and build a table. Surname, stated place if any, shared cM, tree quality. Watch what repeats. You are looking for a parish name that keeps surfacing across unrelated trees.

Now go to the records. Griffith’s Valuation, formally the Primary Valuation of Tenements, was published between 1847 and 1864 and lists the occupiers of every piece of land and property in Ireland, along with the people they leased from. It is organized by townland, the smallest unit of civil administration in the country, arranged within civil parishes, within baronies or Poor Law Unions. Search your cluster surnames against your candidate parish. Surname distribution in Griffith’s will tell you very quickly whether that parish is plausible or whether you have wandered into a name that is common everywhere.

Workhouse records are where you can sometimes close the last gap. Indoor relief registers, where they survive, record full name, sex, age, trade, marital status, religion, spouse’s name, number of children, the electoral division of residence before admission, the date of admission, and the date of death or departure. That residence field is the prize.

Irish Famine Begins: A Hypothetical Worth Walking Through

Everything that follows is invented. The record sets and place names are real, the researcher and her family are not.

Karen has a great-great-grandmother, Bridget Gilmartin, who married in Brooklyn in 1857 and died in 1894. Every American record gives her birthplace as Ireland. Karen has been stuck for eleven years.

She pulls her AncestryDNA matches above 20 cM, works a Leeds chart, and ends up with a cluster of 23 people that does not connect to any of her known lines. Six of the 23 have trees naming Irish origins. Four say County Sligo. Two of those four say the parish of Drumcliff, in the barony of Carbury. Surnames repeating across the cluster include Gilmartin, Gilmore, and Glancy.

Karen checks Griffith’s Valuation for Sligo, surveyed in 1856 and 1857. The Drumcliff returns show a Hugh Gilmartin occupying property in the townland of Drumcliff South, with Gilmore and Glancy households scattered through neighboring townlands including Cartronmore and Doonfore. The cluster surnames and the valuation surnames are sitting in the same few square miles.

Bridget left before that survey, so she will not appear in it. What Karen needs is a record that places the family in the parish during the Famine itself. She turns to Sligo Poor Law Union. Admission and discharge registers for Sligo Union covering 1848 to 1859 have been digitized and indexed on Findmypast, with a gap between 1850 and 1853.

An 1849 admission for a Catherine Gilmartin, age 19, single, laborer’s daughter, with a residence recorded in the Drumcliff area and a discharge four months later, gives Karen a family in place at the right moment. A sister? A cousin? She cannot say yet. What she can say is that the paper record and the genetic cluster now agree on one parish, and that she has moved from “Ireland” to a set of adjacent townlands she can actually research and eventually stand in.

Irish Famine Begins: Where This Falls Apart

Endogamy will wreck a clean cluster. Rural Irish parishes married locally for centuries, so your matches may be related to each other through four different couples rather than one, and colored groups start bleeding into each other. Clustering methods are known to go sideways in endogamous populations.

Workhouse survival is the other hard limit. Survival of indoor relief registers is poor, and there are documented instances of workhouse registers being destroyed in military operations during 1921 and 1922. Board of Guardians minute books survive far better than admission registers, though they carry fewer names of inmates. Poor Law records are mostly held at county level, with some originals held in Dublin at the National Library and the National Archives, and in Belfast at PRONI. Check the county archive first. Several have digitized and published their holdings.

And a cluster pointing at one parish can still be pointing at the wrong parish. Test it. Try to break it.

Irish Famine Begins: Go Group Your Matches

Group your Irish DNA matches by shared surname or county and see which corner of Ireland they keep circling back to. That is the whole assignment. An afternoon with a spreadsheet and your existing match list, no new purchase required.

If nobody in your family has tested yet, that is the thing standing between you and a townland name. An AncestryDNA kit puts you in the largest match database available, which matters enormously for Irish research because the cousins who stayed and the cousins who left both need to be in the same pool for any of this to work. Sales run several times a year, and September is usually a good month to watch.

The families who left in 1847 and 1848 carried almost nothing with them. No deeds, and in a great many cases not even a surname spelled the same way twice. They carried their DNA, and they had no idea they were carrying the one thing that would survive the fire.

* * *

Author’s Note: I want to be transparent that this content – Irish Famine Begins 1845: Using DNA Matches to Find the Townland Records Never Recorded – was created in part with the help of an artificial intelligence (AI) language model – Claude Opus 5. The AI assisted in generating an early draft of the content, but every paragraph was subsequently reviewed, edited, and refined by me. The final content is the result of extensive human curation and creativity. I am proud to present this work and assure readers that while AI was a tool in the process, the story, style, and substance have been carefully shaped by the author.