Using Ancestry ThruLines as Proof

|

Introduction

The introduction of Ancestry’s ThruLines feature showed the genealogy community the impact that AI can have on research. As one of the most advanced genealogical hinting systems ever created and somewhat independent of document research, impressive predictions about unknown ancestors can be made. It’s also just a prediction and not proof.

Objective

This post aims to demonstrate how to responsibly incorporate ThruLines results into our family tree and a genealogical proof summary, highlighting the importance of disclosing its use as supporting evidence rather than primary proof, and emphasizing the preference for direct, well-documented evidence in drawing conclusions.

Preparation

To follow this example, users will need access to an AncestryDNA test with ThruLines enabled. If ThruLines is not currently enabled, refer to Ancestry’s guidance for instructions. Two individuals will be provided for this case study, but you are encouraged to replicate this with your own research.

For the best results, identify two example ancestors who are labeled “potential ancestors” (green color):

  1. An ancestor with no additional children listed beyond your direct line. My example will be Maria Catharina Wolf.
  2. An ancestor with several children beyond your direct line. My example will be Eldred Bentley.

These contrasting examples can help illustrate the incredible benefit of ThruLines, while also highlighting the risks and limitations of using ThruLines as evidence. While using your own research is recommended for a thorough understanding, it is not strictly necessary.

Maria Catharina Wolf's ThruLine

Implementation

For each example ancestor, follow these steps:

A. Explore the ThruLine

  • Click on the suggested ancestor in your ThruLines view to open their ThruLine or click the images above for this case study.
  • Review the number of children listed under the ancestor and note the total number of descendants.
  • For a detailed DNA analysis, take note of the total shared cM values for matches linked to this ancestor.
  • Observe the distribution of DNA matches among descendants of the ancestor.
  • Identify patterns or clusters of matches that may confirm or challenge the relationship.

B. Click on the “Potential Ancestor” to open the “Evaluate” panel

C. Review the sources for the “Potential Ancestor” in the “Evaluate” Panel

  • This panel will display the evidence Ancestry used to suggest this person as a potential ancestor.
  • Carefully examine the sources and records provided in the “Evaluate” panel.
  • Note whether the evidence includes reliable documentary records or simply user-submitted trees.
  • Compile any relevant evidence into your research notes for further analysis.

Review and Verify

ThruLines provides valuable leads for genealogical research, but its suggestions are educated guesses that require critical evaluation. For the examples provided, here’s how to approach the verification process:

Example Analysis: Maria Catharina Wolf

Details: ThruLines suggests Maria Catharina Wolf as the test taker’s ancestor with only one child listed (the test taker’s line) and cites three user-submitted trees (one private).

Evidence: There is no easy way to find evidence for her beyond continuing to research my last known ancestor. The ThruLines tool did not add any evidence.

Assessment: None of the cited trees include documentary evidence to confirm the connection. The suggested relationship remains speculative.

Actions:

  1. Do not add Maria Catharina Wolf to your tree at this time.
  2. Record the suggestion in your notes for future research:

“ThruLines suggests her mother’s name is Maria Catharina Wolf, but no evidence has been found for her. Three trees were cited by ThruLines, none provide sources or documents as evidence for the connection. ThruLines does not list any other possible children as of 1/15/2025.”

Example Analysis: Eldred Bentley

Details: ThruLines identifies Eldred Bentley as an ancestor with four children and a total of 19 descendants who have taken tests at Ancestry. Approximately half of these descendants are in the test taker’s direct line. Several trees are listed, and many include documents.

Evidence: In addition to investigating Eldred Bentley, researching the listed children could yield documentary evidence connecting the siblings, specifically the direct line ancestor, Mercy Bentley, to one of her potential siblings. Importantly, ThruLines is suggesting a sister with her married name. I will want to take note of this woman and research her with both her maiden and married names.

In reviewing evidence for these new possible family members, I discover a will for Eldred Bentley in 1843. It lists these children, but also son Eldred Bentley, daughter Louisa Carr, and another daughter “Mary Nichols Bentley.” I also find an 1855 New York Census with “Mercy Bentley” living in the household of “her mother,” Lydia Griswold, next to Eldred Bentley and Louisa Carr—the remaining two children of the “potential ancestor” who are not yet shown in ThruLines. Mercy Bentley is listed in this census with “(Fuller)” written next to her name as an afterthought. Research in deeds shows Mercy S. Ralph and Jemima J. Jeffords relinquishing the lands deeded to them by Eldred Bentley in 1859. This deed is recorded in 1865 next to one of John and Cecil relinquishing their own lands inherited from Eldred. The deed indicates that they moved to the state of Illinois.

Lydia Griswold’s second marriage is also revealed through online records. There is a deed dated 1850 made out to her that names her as the “Dower” and makes reference to the will of Eldred Bentley, drawing a map of how he divided his land between his three sons (all named) and the portion left to his wife. A biography of a Daniel Griswold connects him to Lydia when it mentions the death of Daniel’s wife in 1844 and says that afterwards “he married a Mrs. Bentley.” A cemetery return says he died in January 1855, explaining why Lydia was the householder in the NY Census that year.

As for Lydia’s own ancestry, an old story from a 1901 county history book features Mercy and also details Lydia’s connection to the Niles family:

“Nearly a mile west of Olds’ Corners, on the Cattaraugus County line, and on the old Chautauqua Road, in a little log cabin, lived one Eldred Bentley, Jr., with his family, among whom was a daughter named Mercy, a simple-minded girl. Eldred had married a daughter of one John Niles, whose wife had succumbed to the hardships of life, and the old man was then making his home with the Bentley family.”

Actions:

  1. Add Eldred Bentley and his family to your tree with clear notes documenting the existing evidence and gaps:

“ThruLines suggests Mercy’s parents were Eldred Bentley and Lydia Niles. Good evidence supports Mercy as a daughter of Eldred Bentley and Lydia. Eldred’s will in 1843 lists all the children found in ThruLines and two additional children, Eldred and Louisa. The 1855 NY Census shows Mercy living with mother Lydia Griswold near apparent siblings Eldred Bentley and Louisa Carr. Lydia’s maiden name is not confirmed, but an old story suggests her father was John Niles, who lived with the Bentleys after John’s wife died. I do have a ThruLine for John Niles but have not looked further into it. Lydia marries Daniel Griswold after 1844 and he dies Jan 1855, leaving her widowed again. ThruLines lists children John, Jemima, and Cecil Niles as of 1/15/2025. The sons moved to Illinois according to the deeds and should be researched further. Look for a John Niles in Chautauqua County.”

  1. Add citations to all of the documentary evidence that was gathered:
    • Eldred Bentley’s Will
    • 1855 NY Census
    • Deeds for Lydia, Mercy and Jemima, the sons, etc.
    • The story from the 1901 county history [look up the original book instead of just the webpage]
    • Biography for Daniel Griswold
    • Cemetery return for Daniel Griswold
  2. (Optional) Research additional descendants of Eldred Bentley on the other family lines, specifically Eldred and Louisa. By adding them into a DNA-connected tree, descendants of these families will more easily appear in the future.

Ethical Considerations

When using ThruLines in genealogical research, it is crucial to recognize its role as a tool, not a source of proof. ThruLines helps gather potential connections for further investigation but does not provide conclusive evidence of relationships.

For example, while my notes may reference ThruLines in guiding me to Mercy’s family, the actual proof of her relationships will come from the documentary sources listed above. Any DNA evidence I cite must stand independently, supported by detailed analysis and correlation with traditional records. At this time, I have not looked into the quality of the DNA matches, but I noticed they were all very small. Regardless, the documentary evidence stands on its own very well.

ThruLines serves as an AI-powered hint, and in many cases, I will not need to disclose its use in my final write-up. However, if I cannot find direct evidence of a relationship, I might choose to disclose the ThruLines suggestion as a lead, clearly stating that it is not proof. Transparency in such cases ensures that the limitations of the evidence are communicated to others reviewing my work.

Additionally, caution must be exercised when adding individuals to a family tree. Incorrectly entered relationships can propagate errors that affect both personal research and shared trees. As more people copy the information, they can become impossible to correct. Always verify information thoroughly before making additions to avoid spreading inaccuracies.

Benefits and Limitations

ThruLines can be a valuable tool for genealogists, offering unique advantages while also presenting certain limitations. Understanding both is essential to using this feature responsibly and effectively.

Benefits

  • Streamlines Research: ThruLines collects DNA matches and tree data into a visual, organized format, helping you quickly identify potential relationships and research leads.
  • Suggests New Connections: By analyzing DNA data and family trees, ThruLines can highlight ancestors or relatives you may not have considered, uncovering paths that simple searches might miss.
  • Identifies Clusters of Matches: The tool groups DNA matches into shared descendants, making it easier to recognize patterns and validate family groupings.
  • Time-saving: ThruLines aggregates evidence from user-submitted trees and matches, saving time compared to manually searching through individual DNA match lists and tree data.

Limitations

  • Reliance on User-Submitted Trees: The accuracy of ThruLines is only as reliable as the trees it uses. Errors in user-submitted trees propagate into inaccurate suggestions.
  • No Source Verification: ThruLines does not verify the sources of tree data, meaning the “evidence” it compiles may lack documentation or reliability. You will need to independently verify everything in ThruLines.
  • Overgeneralization by the Algorithm: The AI powering ThruLines may group individuals with similar names or locations into families, even if there is no direct evidence to support the connection.
  • Not Proof: ThruLines provides hints, not proof. All suggested relationships need to be critically evaluated and confirmed through traditional genealogical methods.
  • Limited to Available Data: If few matches or trees are connected to an ancestor, ThruLines may offer incomplete or inconclusive suggestions.

Takeaway Tips

  • ThruLines is a Starting Point, Not Proof: Use ThruLines to generate leads and potential connections, but always verify these suggestions with documented evidence before adding them to your tree.
  • Evaluate the Quality of Evidence: Examine the sources behind ThruLines suggestions, such as user-submitted trees, and determine if they are reliable and well-documented.
  • Cross-Check DNA Matches: Use shared cM data and compare it to the Shared cM Project or other tools to ensure the suggested relationships align with expected DNA patterns.
  • Document Your Findings: Keep detailed notes on what ThruLines suggests, including its limitations. Record any evidence you uncover for or against the connection.
  • Approach with Caution: Be aware that the algorithm can group unrelated individuals together due to similarities in names, locations, or incomplete tree data.
  • Avoid Propagating Errors: Do not add individuals to your family tree based solely on ThruLines suggestions, especially when they are not supported by strong evidence.
  • Be Transparent in Your Research: If you use ThruLines to guide your research, disclose this in your notes when appropriate, especially if no conclusive evidence is found.
  • Stay Open to Collaboration: Reach out to DNA matches highlighted by ThruLines to exchange information and potentially uncover new records or insights.