Citation of Widow’s Pension images posted on WikiTree

Hello, Brian. Thanks for providing everyone a superb example of how the convenience of online materials can create really complex citations!

You've captured all the essential information.  EE can suggest a few tweaks for clarity.  Immediately below, I am dividing the citation into its three natural layers to more clearly illustrate the points. Alterations are flagged in red (with highlighting for the pieces of punctuation that might not otherwise be noticed):

Layer 1

"Dependent Widow’s Declaration for Pension," rehearing, 23 May 1899, filed with Ann L. Lane, widow’s pension application no. 618,465, W.C. certificate no. 532,473, pension no. XC 2,672,697, service of James R. Lane (Pvt., Co. F, 156th New York Infantry, Civil War);

Layer 2

PDF, posted at "James Riley Lane US Civil War Documents," WikiTree (https://www.wikitree.com/photo.php/4/43/James_Riley_Lane_US_Civil_War_Documents-5.pdf#page=7 : accessed 26 July 2024), page 7;

Layer 3

digital images provided by The Horse Soldier Research Service, http://www.horsesoldier.com, 26 February 2019, without citation of file, series, or record group; typically, such files appear in Record Group 15, National Archives, Washington, D.C.

As explanations:

Layer 1: 

The title of the document should be in quotation marks, given that you are quoting it exactly. Boldface is not standard.

Your draft citation (a) places a semicolon between the pension number and the person to whom it applied; and then (b) places a comma between the last detail for the pension and the first detail for the online site.  This should be reversed. In citations, as in ordinary writing, commas separate simple items in a series, while semicolons mark a major break between items in a series that already have internal commas. (See EE4 §2.71 and 3.18 at the top of p.115,)

In a layered citation, the semicolon marks the break between layers, because we are shifting from one thing to the other. When citing online material, the semicolon indicates that we are shifting from (a) the original document ID to (b) the website ID, then to (c) whatever other outside information further identifies the material. As the draft citation was written, your divisions were this:

Dependent Widow’s Declaration for Pension, Rehearing, 23 May 1899, filed with Ann L. Lane, widow’s pension application no. 618,465, W.C. certificate no. 532,473, pension no. XC 2,672,697;

service of James R. Lane (Pvt., Co. F, 156th New York Infantry, Civil War), digital images provided by The Horse Soldier Research Service, http://www.horsesoldier.com, 26 February 2019, ...

but "service of James R. Lane ..." is not part of the identification of "The Horse Soldier Research Service" or its failure to appropriately identify the materials it provided to you.

Layers 2 & 3:

For an Evidence Style citation, we would switch these two blocks of information.  With image documents, if we cite the original document first, then the order is this (EE4  §3.14):

• original document;
• website that publishes the images just cited in Layer 1;
• any additional information about the source provided by the website or acquired from elsewhere.

The bigger issues here is that you are merging two different situations:

1. You, as the creator of the site, are trying to provide (in the new Layer 3) all the background information that you know the user needs, but
2. You are also citing the record as though you are the user, who "accessed" this document at your web page on a certain date.

The differences between these two functions matter.  You, as the creator of the site, know information that the user does not; that's what you put in the new Layer 3.  But how would the user who accesses the document at its cited URL know this background? Where would the user find this information in order to include it in their citation?

To separate these two issues:

1. You, as the creator, need to annotate the document to state the information that now appears in Layer 3.  (Indeed, when posting a PDF of a document online, "best practice" calls for adding an annotation that fully identifies that document.)
2. The user would then read your annotation and add that to Layer 3 of their citation—or copy your suggested citation.

Layer 2:

The bridge words "imaged in" that you added between the original document (Layer 1) and the website (the new Layer 2)  does not clearly identify their connection. Users of the document aren't dealing with a series of loose images provided at a website such as FamilySearch.  Someone has created a page titled "James Riley Lane US Civil War Documents" at which you posted a PDF.

BOTTOM LINE:

If you add the annotation to the document, then a full citation to it might be this:

"Dependent Widow’s Declaration for Pension," rehearing, 23 May 1899, filed with Ann L. Lane, widow’s pension application no. 618,465, W.C. certificate no. 532,473, pension no. XC 2,672,697, service of James R. Lane (Pvt., Co. F, 156th New York Infantry, Civil War);  PDF, posted at "James Riley Lane US Civil War Documents," WikiTree (https://www.wikitree.com/photo.php/4/43/James_Riley_Lane_US_Civil_War_Documents-5.pdf#page=7 : accessed 26 July 2024), page 7; citing "digital images provided by The Horse Soldier Research Service, http://www.horsesoldier.com, 26 February 2019, without citation of file, series, or record group; typically, such files appear in Record Group 15, National Archives, Washington, D.C."