Image through index

A couple of questions on this citation. First, while the book itself has the name "Kirkebok" embossed on the cover, the archive calls it a klokkerbøker [digital name], Kirkebøker [on label], and Klokerbok [on label]. They all are very close when you translate, it's church book or the watch book. I used the name embossed on the cover and don't reference the other names directly.

Next layer 1 is fairly straight forward, the only catch is that the entire book is section E. In earlier parish records birth, marriage, death ... were all in one book with sections, that changes in this time frame to a specific book per event. Even though there are not separate sections in this record i still put in the section label.

Layer 2 represents a HUGE advantage for Norwegian genealogists, the Digital Archives. BMD and census records are usually scanned and indexed ALL UNDER ONE SITE!! Yipee. There are additional records also scanned and indexed. Not everything is scanned, but for BMD that's nearly 100%. Not all are indexed. So in this case layer 2 points to the digital index, and the permanent ID for the event. While the web address may change it's unlikely they will change the ID for the event.

Layer 3 is the image source. It also has the archive reference. I did not break the reference down into components, but did add the period reference. While you can get to the image from the index (using the scanned button), i've included the link in the citation

Given that most will find the source through the index, that is layer 2, it feels wrong for the index to be level 3. 

Vår Frue Parish (Trondheim), “Kirkebok [church book]”, 1922-1949, section “E Døde [Died],” p 56 (stamped), entry 41, buried 11 September 1930, Kristoffer Trondsen Søberg; indexed in “Digital Archives,” National Archives of Norway, permanent ID pg00000005447190 (; citing Regional State Archives in Trondheim, reference SAT/A-1456/602/L0150, period 602C18/1922-1949,  indexed image (, image 58 of 193.

Submitted byEEon Thu, 04/01/2021 - 10:05


The issue you are wrangling with is this: You are trying to cite two totally different things within one citation census. Those two things are

Your citation begins with the identification of the original church book, its page, and entry of interest. Your citation ends with the URL at which you viewed that page of the church book. But you are separating those two by a reference to a totally different creation: the website’s database abstract.

EE would handle the citation this way (using different colors for each layer):

       1. Vår Frue Parish (Trondheim, Norway), “Kirkebok [church book]," 1922-1949, section “E Døde [Died],” p. 56 (stamped), entry 41, Kristoffer Trondsen Søberg, buried 11 September 1930; imaged, National Archives of Norway ( : accessed 1 April 2021); citing “SAT, Ministerialprotokoller, klokkerbøker og fødselsregistre - Sør-Trøndelag, 602/L0150: Parish register (copy) no. 602C18.”  The archives’ abstract of this burial entry, accessible through the search function on the home page, is at, citing “Permanent ID: pg00000005477190.”

Note a few things here:

  • The location of Trondheim needs to be provided.
  • For Layer 3, where we say that the provider is citing ... [whatever], EE would cite exactly what the archive cites in its header above the image. Note that this includes the fact that the book is a “copy.”
  • The permanent ID is provided on the abstract page, not on the image. That piece of data should be grouped with the abstract.
  • The position in which you placed the permanent ID, between the title of the publication and the publication data, is a position reserved for data that modifies the title of the publication—example: revised edition.  The permanent ID number does not modify the title of the publication (i.e., the website title). It identifies the specific item of interest; the "item of interest" field is always positioned after the publication data. As a memory crutch here, just remember that a website is cited like a book: 

Author, Title of Publication, rev. ed. (Place of publication : date), page number or whatever specific item of interest.

  • The image number is not included because you are not citing a specific database to which that image number applies. The archive has many different databases with an "image 58."  (In this case, the "image 58" references the 58th image of that church book; but the URL for the image also provides that data. No citation police would slap your wrist for repeating it, but it's not essential and would be redundant.)

I am about to document many source/citations from church records in the National Archives of Norway like the one seen in this thread. I fully agree with the way layers 1 and 2 are created, but have a question about layers 3 and 4. Isn't it sufficient to have layer 3 state, "citing the Regional State Archives in Trondheim"? Or if more information is desired to add what is shown on the page with the original image, such as, "Reference SAT/A-1455/602/L0150, register no. 602C18/ 1922-1949"? It seems sufficient to give the URL of the document and not have a second URL for the indexed extraction. Also, I wonder if the image number should be mentioned in layer 2? It is the last numbers in the URL, but may not be obvious to many people that it is the image number.

Mike, you ask:

Isn't it sufficient to have layer 3 state, "citing the Regional State Archives in Trondheim"? Or if more information is desired to add what is shown on the page with the original image, such as, "Reference SAT/A-1455/602/L0150, register no. 602C18/ 1922-1949"?

For the source-of-the-source layer, if the website provider identifies the document's reference number within the archives that holds the record, then it would not suffice to just identify the archives. The document's reference number is essential.

You also say:

It seems sufficient to give the URL of the document and not have a second URL for the indexed extraction.

In most cases, yes, we would cite the document that we use, rather than the database entry. However, there are situations in which we might need to cite the database's entry also—as, say, when an error appears therein that we make a point of correcting. Cryptoref asked about citing both the image and the database entry; so our response covered that as well.

And then you ask:

Also, I wonder if the image number should be mentioned in layer 2? It is the last numbers in the URL, but may not be obvious to many people that it is the image number.

If the URL we cite takes us precisely to that image (the specific item), then we would not cite the image number because that URL would not have, say, 58 different images to scroll through. If the URL takes us to only to the collection level, then we would need to say which image in that collection contains the specific item we are referencing. In this case, the referenced page is

the 58th image at

but it is not

the 58th image at

Because takes us directly to the image of what we are referencing, we would not then add "image 58."

Submitted byLene Kottalon Thu, 04/01/2021 - 18:04


Klokkerbok does not mean watch book, but rather the clerk's book. In Scandinavia, church books were kept in duplicate, one by the minister and the other by the parish clerk. The clerk's book (klokkerbok) is the copy. 

The mistranslation arises because the word klokker means both bells and bell ringer. A parish clerk used to be called a klokker, because ringing the church bells was his responsibility. 

Lene D. Kottal 

Submitted bycryptorefon Sun, 04/04/2021 - 08:35

Thank you. I knew it was two different things, but didn't think about making it into two different sentences. Should have occurred to me, but that's why I'm still learning.



Submitted byEEon Sun, 04/04/2021 - 09:39

David, we're all still learning. Technology keeps offering new wonders and new challenges. 

Submitted bycwhermann28on Thu, 05/27/2021 - 19:22

Citing the image of published work from a database with images located on Ancestry.

In this instance the source of the index is a published work of 21 volumes published over a span of two decades consisting of details extracted from vital records for the state of Rhode Island. of extracts from another source.   I have reviewed EE9.6 and 9.32-34 along with chapter 12 in attempt to understand how to create this given the source of the index is a published work which is a compilation of extracts from town records.

The publication is also available on other digital library sites like Internet Archive and Hati Trust, but the user can only search on a give volume so the index on Ancestry acts like a nice index of all 21 volumes.  I decided to model like a publication with online images,  using the database with images at Ancestry as the second layer.  Here is what I came up as first draft.  I am not sure if I need the way points since the image is found via the database search as apposed to browsable or un-indexed images often found on FamilySearch.

Bibliography: Arnold, James N, compiler. Rhode Island Vital Extracts, 1636-150: First Series, Births, Marriages and Deaths. 21 volumes. Providence, Rhode Island: Narraganset Historical Publishing Co., 1891–1912. "Rhode Island, U.S., Vital Extracts, 1636-1899." Database with images. Ancestry. : 2019-2021.

Footnote: James N Arnold, compiler, Rhode Island Vital Extracts, 1636-150: First Series, Births, Marriages and Deaths, etc., 21 vols. (Providence, Rhode Island: Narraganset Historical Publishing Co., 1891–1912), vol.4, p. 21, entry for Nicholace Mosher; citing vital records for the State of Rhode, Island; accessed through "Rhode Island, U.S., Vital Extracts, 1636-1899," database with images, Ancestry ( : downloaded 25 May 2021) > Vol. 4 > Newport County > Births, Marriages, Deaths > image 612 of 691.

Short Footnote: Arnold, Rhode Island Vital Extracts, 1636-150, vol.4, p. 21.

cwhermann28, EE 9.6 (online images of original county records offered in a database) would not be applicable to what you are using (online images of a set of published books).

Ditto for EE 9.32 which covers online database entries without images of anything

When a published book "talks about" a type of record or provides details from a type of record, that's not the same thing as citing those records itself. We're simply citing the published book.

Chapter 12 "Publications" is the appropriate one. That's what you're citing: images of a publication. Specifically, a series. That's covered by EE 12.84 "Series & Occasional Works." 12.84's discussion provides the basic citation you'd use for layer 1. Then, because you are using images of the book provided by Ancestry, your Layer 2 would cite Ancestry's database, etc.  That's what you've done above, so you're on target.

Incidentally, when a new query is tacked onto a "retired" thread, this website's system does not alert me. My apologies for not seeing this sooner.

Submitted byKristinaCleveron Tue, 02/28/2023 - 10:32

I am diving back into my Norwegian research and want to ensure I get this right. Other than not being able to read one of the words for the title in layer one, I think I have layers one and two in good order.

Layer one: Mo Parish (Telemark County, Norway), "No.5 Ministerialbog for Moe praestegjeld i ovre Thelemarkens [?] Provstie, Begyndt (beginning) 12 Mai 1844, Sluttet (ended) 31 December 1865," section "D, Begravede og dødfødte [buried or stillborn]," p. 344 (penned), Vetle Nerisen, 23 January 1860 (death);

Layer two: imaged, National Archives of Norway, Digitalarkivet [italics] ( accessed 27 February 2023);

Layer three: citing "Mo kirkebøker, SAKO/A-286/F/Fa/L0005: Ministerialbok nr. I 5, 1844-1864."

My question is for layer three. Studying the information available tells me this:

the physical location (Oppbevaringssted): Statsarkivet i Kongsberg (archive in Kongsberg)

archive (Arkiv og arkivdel): A-286: Mo kirkebøker

series (Serie og underserie(r): F/Fa

piece/folder (Stykke/mappe): Kirkebok

source type (Kildetype): Ministerialbok

protocol no. (Protokollnr./tidsrom): nr. I 5 /1844 - 1864

I would prefer to spell out the physical location prior to the title for clarification rather than just the acronym SAKO. Would it be preferred to include the title as shown on the website (option 1) or would it be sufficient to start the title after the elimination of SAKO (option 2)?

Layer three option 1: citing Statsarkivet i Kongsberg, "Mo kirkebøker, SAKO/A-286/F/Fa/L0005: Ministerialbok nr. I 5, 1844-1864."

Layer three option 2: citing Statsarkivet i Kongsberg, Mo kirkebøker, "A-286/F/Fa/L0005: Ministerialbok nr. I 5, 1844-1864."

KristinaClever, you’ve done well. If you wish to spell out the cryptic abbreviations in Layer 3 (the “citing ….” layer), that would be helpful. EE would use your option 1. Since your intent is to quote what the website is citing in its header/tag above the image, it would be better to cite the whole tag rather than start in the middle of it.

Incidentally, in tracing your steps, I noticed two wee things:

  • Unless I’m looking at a different title page from the one you used (I’m using image 2), the book’s ending year is 1864 rather than 1865, no?
  • It would be helpful if your Layer 2, immediately after the parenthetical publication data, cited the image number at which Layer 1’s record is found.

Thank you. I caught the typo at the end of layer 2 and have added the image number.

Due to the challenges with Norwegian research, I am considering adding the farm name to layer 1. The farm name is typically included in records and helps separate same-named individuals. I have also found that a few of my Norwegian relatives used their farm name as a surname after immigration. 

Vetle's son Neri Vetlesen immigrated between 1860 and 1864. Along with the multiple spelling variations of Vetlesen, Neri used the surname "Vadder" (the farm where he was born). The surname was recorded as Wilson in a later generation that created my 37-year brick wall.

Kristina, when we use a source—particularly a complex one—we may add whatever information we feel would be helpful to the source-identification and source/information-analyses. Editors at a later date, if they become involved, may  trim our citations to fit their cost-saving models; but if we are thorough in our citations, we will never face the problem of those editors wanting information we did not record.

Wilson = Vetlesen! That's one I've not encountered. In retrospect, it's understandable; but it would indeed create a roadblock, until recognized.


Submitted byKristinaCleveron Tue, 05/16/2023 - 08:35

I have a follow-up question regarding the preferred URL for the digital image.  There are several references provided in when you click on the "i" on a digital image. These are the English translations.

Permanent image ID
Quick Link
Permanent image-link
Search transcribed source - Born and baptised
Search transcribed source - Deceased and buried
Search transcribed source - Married

The URL from the address bar, which I have been using, is not included in the References section. Although the URLs I have been using do bring me back to the desired image, I am concerned that they may not be stable URLs ( This likely means I need to edit all the citations I previously created. My question which of these URLs should I include in the citation? 

The Quick Link includes the Permanent image ID, so I am leaning toward that one. In my communications with another researcher, he always references the ID number. (

If I click on the Permanent image-link, I immediately see a big problem. While it shows the image, it does not include any references to get me back to the original source. (