Are the Digital Image Collections of the Published Pennsylvania Archives books at Hathi Trust and Fold 3 Considered Complete?
Myrt asks Perplexity to evaluate whether all pages from the 138 books are included when doing AI searches of certain online digital image collections.





Last April, in Lost in Pennsylvania? Ol’ Myrt here briefly described in what order to access colonial era research across ten series (138 volumes) of the Published Pennsylvania Archives, not to be confused with the Pennsylvania Archives building.
Bogged down by the mishmash of entries concerning my ancestor Conrad Weiser, I intend to ask AI to canvas the Published Pennsylvania Archives and sort Conrad’s entries in chronological order.
Before asking, I want reassurance that the two publicly available scanned image collections I’m working with are complete - namely HathiTrust and Fold3. (I had previously noted that FamilySearch had an incomplete set, and I didn’t want to use the search engine at Ancestry.)
Myrt’s Prompt
In your research of the published Pennsylvania Archives at HathiTrust and Fold3, are there any missing digital images or do you consider the digital image collections complete?
Perplexity’s Response
Based on available information about the digitization of the Pennsylvania Archives at HathiTrust and Fold3, the collections are extensive but not necessarily complete for every possible image or document.
Scope of Digitization:
The Pennsylvania Archives are published in 10 series totaling 138 volumes, covering a wide range of colonial and early state records. These have been digitized and are available online at HathiTrust and Fold3, with full-text search capabilities and digital images of the published volumes[1][2].
Fold3 specifically hosts the most complete and indexed collection of these published volumes online, making it a robust resource for researchers[2].
Completeness of Digital Images:
The published volumes themselves are digitized in full, meaning all pages of the printed books are available as digital images[2].
However, the term "complete" only applies to the published volumes themselves, not to every original manuscript, map, or supporting document referenced within them. Some original records, maps, or unique documents may not be digitized or may only be available as transcripts within the published volumes[1][2].
Access to Original Manuscripts and Supporting Materials:
The digitized collections primarily provide images of the published transcriptions and compilations, not always the original handwritten documents or every supporting item referenced in the text[2].
For some record groups, original manuscripts and supporting materials may be available elsewhere, such as the Pennsylvania State Archives or other repositories, but these are not always included in the HathiTrust or Fold3 digital collections[1][3].
Ongoing Digitization Efforts:
The Pennsylvania State Archives have been actively digitizing microfilm and photographic collections, but these efforts are separate from the published Pennsylvania Archives volumes and focus on original records and images not included in the published series[3].
These additional digital images are made available through platforms like PA Power Library and PA Photos and Documents, not necessarily through HathiTrust or Fold3[3].
Conclusion:
The digital image collections of the published Pennsylvania Archives at HathiTrust and Fold3 are complete in the sense that all pages of the published volumes are digitized and available[1][2].
However, not all original documents, maps, or supporting materials referenced in the archives are included in these digital collections. For those, researchers may need to consult additional sources or visit physical archives.
1. https://librarycompany.org/portfolio-item/pennsylvania-archives/
2. https://macysgenealogy.com/pennsylvania-archives-online-resources/
3. https://thecrowleycompany.com/pennsylvania-state-archive-digitizes-collections/
Myrt’s Thoughts
I knew some of the supporting documents, such as original muster rolls, may no longer exist, or may have disintegrated over time. That’s what happens with loose papers in folios, even at archives.
Dealing with finding the original document abstracted in the Published Pennsylvania Archives wasn’t my question. YES, competent genealogists always prefer looking at the originals from which a book of extractions is published. Keep in mind, though, in this case, even the venerable National Society Daughters of the American Revolution accept the Published Pennsylvania Archives as a reliable source, since the state of Pennsylvania itself produced the book series.
To be more precise, I should have composed my prompt to include “every page, from every volume in each of the ten series.” So, perhaps this prompt would suffice: In your research of the 138 volumes of the published Pennsylvania Archives books at HathiTrust and Fold3, are there missing digital images of pages? Do you consider the digital images of this large book collection complete with every page, in every volume?
Summary
Now that I’m confident Perplexity will be able to search every page from every single one of the 138 books in the Published Pennsylvania Archives, my next prompt will be to have it look at all entries for my ancestor Conrad Weiser, Mr. Weiser, *Wiser, and Weiser to see what happens. I’ll be sure to specify at least the following. It’s date order I’ll need first, to construct a timeline of interaction between Weiser and the provincial government.
Date Order
Name referenced
Series, Volume and Page Number
Happy family tree climbing!
Myrt :)
DearMYRTLE,
Your (not so) retired friend in genealogy.
I have heard at least 3 talks given by DAR members who say that before the 1970s sources were not vetted or checked for accuracy. I have even asked other DAR members and they say that is the case. So, Point taken that you still need to verify sources. This is great. I have the same question about the PA Archives as my mom's father was from Pennsylvania and have ancestry there from the Revolutionary war and before (Menonites: Grossman) and German Lutherans (Runkle) , and possibly Quaker (Sankey) or Presbyterian (though the Quaker lineage is under dispute.) I can remember struggling with the physical copies at San Diego State University Library when there were no indexes in the year 2000.
Shame all the documentation in those books was tossed by what was at the time the State Library staff. No way to double check it.