5 things I learned researching the Drouin Collection on Ancestry and Généalogie Québec

In this post, I share what I’ve been learning about record collections outside my normal areas of research. I’m accredited in the provinces of Western Canada (BC, AB, SK, and MB). If you’ve been researching a while, you may know that every time you cross a border, the research changes. It’s a whole different world in Ontario, where records predate Confederation (1867), way back to when Ontario was Upper Canada. It’s a big change from my current province of Saskatchewan, once the Northwest Territories, which reluctantly joined Canada in 1905. My nearest research equivalent to Ontario is British Columbia (1871), but I have yet to see an available B.C. collection that rivals Ontario church and land records. Like any genealogist, I’m a specialist in one area and a newbie in everywhere else. The fun in genealogy is the limitless opportunities to learn.

I’ve been researching the roots of an Irish Catholic family who immigrated to Canada before both Confederation and the Great Famine¹: Richard Donovan (1816-1892) and his wife Mary Ryan (1821-1889), who lived in Stormont, Dundas, and Glengarry Co., Ontario. I haven’t yet learned when they arrived, but it was before 1838. There are a number of delightful genealogical brick walls with this case, namely:

• multiple online trees showing Richard’s father as James Donovan, but none with solid evidence backing that relationship
• Nil 1842 census for the Stormont, Dundas, and Glengarry Counties²
• Nil passenger lists³
• Nil grave marker information
• Nil vital statistics information for Ontario before 1869
• incomplete or missing Catholic registers

Richard and Mary had eleven known children. As one does, my first task after building the family with census records was to locate information about their births. Since there are no Ontario vital records for births 1838-59, the best records to locate for evidence of birth are baptismal records.⁴ Richard and Mary were Irish Catholic, a blessing for a genealogist, because this meant I could dig into the wonderful Drouin Collections.⁵

According to Généalogie Québec, “The Drouin Collection Records is a collection of images of parish registers (baptisms, burials and marriages) as well as of other documents of historical and genealogical significance. It covers all of Quebec and French Acadia as well as parts of Ontario, New Brunswick and the North-Eastern United States.”⁶

The Drouin Collection is available with Ancestry and Généalogie Québec. Thanks to Ancestry, I have access to their databases,⁷ but for this work, I also signed up for access to Généalogie Québec.

No. 1: The Drouin Collection holds records for Catholics outside Québec

Don’t be confused. Généalogie Québec holds records for areas outside Québec, such as Acadia, Ontario, and the U.S. In particular, the collections for Ontario are excellent.

No. 2: Drouin Collection records could be sorted by both diocese and parish

I do love me a good database. Setting aside the search engines for a moment, I saw that Généalogie Québec could arrange records through parish and diocesan folder hierarchies that may contain different records. In other words, if you can’t find your records under the place name, try looking for the diocese. In the below example, the folders show both an “Alexandria,” and a folder for “Alexandria, Diocèse d’ (Catholiques),” or the Catholic Diocese of Alexandria. To be clear, these are not the same.

I can’t tell you how many hours I expended looking for the records in the wrong place. It turns out the records I need, for this Cornwall area family, were added to the Drouin Collection in 2025, and filed under a separate folder path: Ontario/Alexandria, Diocèse d’ (Catholiques).⁸ From a systems view, I understand why they didn’t refile 91K records, but from a researcher’s view, it was a tad confusing.

No. 3: The Drouin Collections at Ancestry and Généalogie Québec are not the same

As genealogists, we are always looking for ways to save money. Before paying another penny, we exhaust the free resources, such as FamilySearch.⁹ Trust me when I tell you I scoured the records at Ancestry before thinking: I wonder if Généalogie Québec has different records?

Let me show you with an example.

Where was the 1838 baptism of John Ryan Donovan?

I was looking for the baptism of Richard and Mary’s first son, John. From the 1851 and 1861 censuses, I knew there was a “John Donovan,” who may have been a son.¹⁰ A search on Ancestry had nil results, and no census told me which Catholic parish the Donovans frequented. Over time, I collected clues about location, eventually finding both Alexandria and Cornwall to be of interest.

I need to pause here to say that, as genealogists, we become expert at finding records because we understand how they are organized. Every site files and categorizes records differently, and each structure can lead to confusion. In Ancestry’s case, manual searches are first by alphabetical place name – “A” for Alexandria – and then by year.¹¹ At Généalogie Québec, manual searches are done by full place name, then decade, then year. If you don’t know how the sites are organized, you can miss important records.

On Ancestry, the records for the years I wanted at Alexandria are 1836-56. This seemed promising, but the records are for St. Finnans Cathedral. I didn’t know it then, but the Donovans worshipped elsewhere. For the City of Cornwall, the records for St. Columban’s are grouped 1834-1838 and 1842-1857. Not only are there four years missing, but the first set holds only 1834-1835. Effectively, this means there are seven years of St. Columbans record not available on Ancestry.

On Généalogie Québec, you can find the 1838 register for St. Columbans using this path: REGISTRES/ Ontario/ Alexandria, Diocèse d’ (Catholiques)/ Cornwall (St-Columban’s)/ 1830/ 1838.

This one record was the key to a cascade of others. It proved the “John” listed in the early censuses was a son, and it led to locating the correct burial record for Richard in 1892. Without John, whose movements were integral to understanding the family cluster, much of the story of the Donovans would have been obscured. (As an interesting aside, connecting this first son to the family initially messed up Ancestry’s DNA ThruLines, as the algorithm adjusted to calculating if my tree jibed with the research of others. I waited a nerve-wracking two weeks before the ThruLines reappeared.)

No. 4: Only rely on search engines when they have positive results.

I knew this already but it’s worth repeating. No search engine is perfect. Not every Drouin collection record is indexed. Perhaps because they are Ontario-based and not Québec-based, I had uneven results using Généalogie Québec’s search tools: Genealogy Quebec search and LAFRANCE (Baptisms, burials and marriages).¹²

For example, here is a search for Bridget Donovan, second child of Richard and Mary Donovan, born 29 Jan 1841 and baptized at St. Andrews West, Stormont Co. It was a good test of Généalogie Québec’s search tools because Ancestry’s algorithm found Bridget’s baptism.

For the Genealogy Quebec tool, the results show a birth for a Bridget Donovan, b. 1842, to John Donovan and Mary ONeil, in Ste-Catherine-de-Fossambault, Québec. It’s tempting but this isn’t our Bridget Donovan.

Here is a search result for Bridget Donovan in the LaFrance search tool. There appears to be no place for St. Andrews West, Stormont. Confusingly, there is a St. Andrew in New Brunswick, which is where this Bridget Donovan was baptized, daughter of John Donovan and Mary Acock.

The finding is that, while this record is in the website’s Drouin Collection records © 2026, it’s not currently findable through either of the big search tools.

No. 5: There may be missing parish records.

Even when you’re reading every page, you may not find the record you seek. You may have located the right location, parish, and year on either Ancestry or Généalogie Québec and you may find the record you need is missing. Missing as in: not digitized.

For example, we know that John was baptized in 1838 at St. Columban’s, Cornwall. This suggests that Richard Donovan married Mary Ryan before 1838. From other records, we know Mary Ryan’s parish was St. Andrews West, Stormont. I would dearly love to locate that marriage record, which could be St. Andrews West or St. Columbans.

Sorry, another aside before we go on. In genealogy, we can’t assert there are no records. We say there are nil records, which means not in location. We say nil instead of no because really, we don’t know if there are no records. There are many reasons why we may not be able to find records, from burning courthouses to natural disasters to paper-only records, but we cannot say a record was never created. The fun of genealogy is accessing a previously unknown set of records. Earlier in this blog, I mentioned the Cornwall area Catholic records were added to Généalogie Québec in 2025. Therefore when I explain in this section there are missing records, I’m not saying they are missing forever. There is always hope someone will find them and digitize them.

I found for St. Andrews West, there are six months missing in the register for 1836. In Ancestry’s “Ontario, Canada, Catholic Church Records (Drouin Collection), 1802-1967,” path: S/ St Andrew`s West/ ALL/ 1836-1856, there is on the first image this marker (see left), which indicates the register’s dates: “Jun 10 1836 au (to) Dec 28 1856, St. Andrews West, Co. Stormont, Ontario.” Checking the previous range, 1802-1835, confirms there are no entries for Jan-Jun 1836.

I checked the same record on Généalogie Québec, which confirmed the first image was a duplicate of this page (but missing the register note).

There are also a number of missing registers for St. Columbans, Cornwall, from about 1830-1850.¹³ With these record gaps, I can’t confirm if Richard Donovan married Mary Ryan before 1838 in Canada or Ireland, nor where they married in Canada (St. Andrews, St. Columbans, or somewhere else). All I can confirm is they did marry, thanks to a range of baptism, burial, and census records.

Final thoughts

Researching this case has given me new insights into the Drouin Collection. I’ve learned the collection holds records for Catholics outside Québec; the records could be sorted by both diocese and parish; the collections at Ancestry and Généalogie Québec are not the same; only to rely on search engines when they have positive results; and that there may be missing parish records. I’ve been given an excellent opportunity to stretch my skills, compare the records across two databases, and establish the hitherto unknown origins of this early Irish Catholic family. I have a new appreciation for Catholic church records, and I hope someday to see the missing registers somewhere. Anywhere.

The title image for this blog was created by the author using ChatGPT Plus on 26 Jul 2026. All the words were created by the author using her own little grey cells.

Thank yous

Thanks this week go to my client Jeff for trusting me with his family’s history and giving me limitless time to talk genealogy with him.

References

1. “Great Famine (Ireland),” 14 July 2026, at 04:05 (UTC), wiki, Wikipedia: https://en.wikipedia.org/wiki/Great_Famine_(Ireland), accessed 25 Jul 2026. ↩︎
2. For a list of what did survive, see “Census of 1842, Canada West, Districts and sub-districts,” 30 Jul 2025, archives and museums, Library and Archives Canada, https://www.canada.ca/en/library-archives/collection/research-help/genealogy-family-history/censuses/pre-confederation/1842-west/districts.html, accessed 26 Jul 2026. ↩︎
3. Generally, passenger lists for Eastern Canada begin 1865 although research shows an alternative research collection with the HBC Archives. See “Passenger lists 1865 to 1935”, date, archives and museums, Library and Archives Canada, https://www.canada.ca/en/library-archives/collection/research-help/genealogy-family-history/immigration/passenger-lists-1865-1935.html, accessed 26 Jul 2026, and my post Western Canadian migration before the CPR: Tracing the voyage of the Princess Royal, London to Fort Victoria (1858-59), 19 May 2024. ↩︎
4. Baptismal records are considered more reliable than census records because they are usually created at or near the date of the event by a reliable witness – a priest in this case – while the birth information on a census was provided by an unknown person, in most cases years after the event. ↩︎
5. To learn more about the Drouin Collections, see the FamilySearch wiki at https://www.familysearch.org/en/wiki/The_Drouin_Collection:_Six_Databases. ↩︎
6. “Drouin Collection: Quebec Parish Records (1621-1940),” undated, archives and museums, Drouin Collection: Quebec Parish Records (1621-1940), Généalogie Québec, https://www.genealogiequebec.com/en/tools/drouin-collection, accessed 26 Jul 2026. ↩︎
7. I have been a member of the Ancestry Canada Board of Advisors since 2018. ↩︎
8. “New Parish Registers — and 25% Off to Access Them!” 24 Sep 2025, blog, Généalogie Québec, https://www.genealogiequebec.com/blog/en/2025/09/24/new-parish-registers-and-25-off-to-access-them/ accessed 26 Jul 2026. ↩︎
9. For example, this database which unfortunately does not have records for Stormont Co.:”Canada, Ontario Roman Catholic Church Records, 1760-1923.” Database with images. FamilySearch. http://FamilySearch.org : 14 June 2024. Catholic Church parishes, Ontario. ↩︎
10. Neither of these censuses recorded relationships, forcing a researcher to infer how people related to one another until verified by other records. ↩︎
11. A place name is the name of a town or city, not the name of the parish. ↩︎
12. Sometimes you get lucky and the algorithm will locate a missing record, as Ancestry did for John’s sister Bridget, but of the eleven total baptisms, I found seven manually. Two remain outstanding. ↩︎
13. Missing records for St. Columbans 1830-50 on GQ: 1830-33 inclusive nil records; 1834 has only Nov-Dec; 1836 has only 14 images; 1838 is missing some months and has only 8 images; the years1839-41 inclusive nil records; 1842 missing records up to 30 May; 1843 nil records after June 23; nil records for 1845. ↩︎