Notes
244 notes
- AncestryProGenealogists, commissioned research journal on the German origin of Charles Johnson, February–June 2020 (unpublished). [PGJ]
- The Trendelburg researcher, transcriptions and family charts (Stammtafeln) for families of Deisel, Trendelburg, Friedrichsfeld, Hofgeismar (Altstadt) and Lippoldsberg, with families at Wehrden, Grebenstein and Schöneberg, and a list of emigrants from Deisel to 1859 (spreadsheet, 15 Oct 2025); spot-checked against the registers, 8 November 2026 (six entries: three agree; the birth dates of Deisel 1835 Nro 199, Hofgeismar Altstadt 1837 Nr. 469 and Friedrichsfeld 1831 Nro 6 differ from the register); emails between that researcher and Jed Johnson, 26 July–8 August 2019 and 24 August 2025–8 November 2026, privately held by the author. [Trendelburg researcher's charts, 2025]
- Klaus Kunze, Ortssippenbücher (HeiKun-Verlag) for Helmarshausen (1521–1950). [OSB]
- Database of births, deaths and marriages for Gottsbüren and the surrounding parishes (“Gottsbüren und Umgebung”). [Gottsbüren database (Dornette)]
- Board for Certification of Genealogists, Genealogy Standards, 2nd ed. rev. (Nashville, Tenn.: Ancestry, 2021). E-book edition (2nd ed., rev.), chapter 1, “The Genealogical Proof Standard.” Elizabeth Shown Mills, Evidence Explained. Chapters 1–3, pp. 17–134. [BCG] [MILLS]
- Roger P. Minert, Deciphering Handwriting in German Documents. [MINERT]
- Blaine T. Bettinger, The Family Tree Guide to DNA Testing and Genetic Genealogy, p. 68. Blaine T. Bettinger, The Shared cM Project, version 4.0 (March 2020), averages and ranges of shared centimorgans by relationship, as presented in DNA Painter’s Shared cM Project tool (dnapainter
.com/ , accessed 6 November 2026). [BETT] [SCP4]tools/ sharedcmv4 - Walter D. Kamphoefner, “St. Louis Germans and the Republican Party, 1848–1860,” Mid-America 57 (1975). [KAMPH]
- Simone A. Wegge, “Chain Migration and Information Networks,” Journal of Economic History 58, no. 4 (December 1998); Simone A. Wegge, “Occupational Self-Selection of European Emigrants,” European Review of Economic History 6, no. 3 (December 2002); Martina Lüdicke, „Verschwunden ist der Strand…“ Studien zur Auswanderung nach Amerika aus Kurhessen in den Jahren 1831–1866, dargestellt am Beispiel des Dorfes Deisel (Kreis Hofgeismar) [“The shore has vanished…”: studies of emigration to America from Kurhessen in the years 1831–1866, shown by the example of the village of Deisel (Kreis Hofgeismar)] Die Geschichte unserer Heimat 14 (Hofgeismar: Verein für hessische Geschichte und Landeskunde, Zweigverein Hofgeismar, 1994). Read in full in a copy borrowed through interlibrary loan, 5 Nov 2026; Walter D. Kamphoefner, “Who Went South? The German Ethnic Niche in the Northern and Southern Hemispheres,” Social Science History 41, no. 3 (2017). [WEGGE98C] [WEGGE02] [LUED] [KAMPH17]
- City of St. Louis, certificate of death, burial permit no. 10619 (register no. 4166), image 1009. [DC03]
- AncestryProGenealogists, commissioned research journal on the German origin of Charles Johnson, February–June 2020 (unpublished). [PGJ]
- Declaration of Recruit, Volunteer Enlistment and surgeon’s form for examining a recruit, 2 Dec 1862, St. Louis. [ENL62]
- Compiled Military Service Record, Charles Gannsen, Battery B., rolls 327 and 762; Charles Gannson (Private, Battery B, 2nd Missouri Light Artillery, Civil War), pension application no. Inv. 749,432, certificate no. 684,507; Case Files of Approved Pension Applications of Veterans Who Served in the Army and Navy Mainly in the Civil War and the War with Spain, 1861–1934; Civil War and Later Pension Files; Record Group 15: Records of the Department of Veterans Affairs; National Archives, Washington, D.C.; digital copy of the file supplied to the author by the National Archives, imaged at https://
charles-research : accessed 30 October 2026. Mail-order copies ordered by the author, first a partial file, order F21-314123835E, submitted 14 February 2017 and delivered electronically, and then the complete file, order F41-523680269E, submitted 8 September 2019 and sent by mail. Charles was a sergeant from 7 November 1863 until 15 March 1865. [CSR] [PEN].netlify .app/ pension-file .html - Working positions of the research, October 2026. [DEC]
- Working positions of the research, October 2026; The surname record: twenty-eight attestations of Charles’s surname in fourteen spellings. [DEC] [SURN]
- Declaration of Recruit, Volunteer Enlistment and surgeon’s form for examining a recruit, 2 Dec 1862, St. Louis. [ENL62]
- Working positions of the research, October 2026; Parish search coverage, 57 places of Kreis Hofgeismar. [DEC] [COVER]
- St. Marcus Evangelical Church (St. Louis, Missouri), marriage registers 1850–1880, images 324–449, image 331, image 355, p. 98, image 55. [WALL]
- Working positions of the research, October 2026. [DEC]
- Charles Gannson (Private, Battery B, 2nd Missouri Light Artillery, Civil War), pension application no. Inv. 749,432, certificate no. 684,507; Case Files of Approved Pension Applications of Veterans Who Served in the Army and Navy Mainly in the Civil War and the War with Spain, 1861–1934; Civil War and Later Pension Files; Record Group 15: Records of the Department of Veterans Affairs; National Archives, Washington, D.C.; digital copy of the file supplied to the author by the National Archives, imaged at https://
charles-research : accessed 30 October 2026. Mail-order copies ordered by the author, first a partial file, order F21-314123835E, submitted 14 February 2017 and delivered electronically, and then the complete file, order F41-523680269E, submitted 8 September 2019 and sent by mail. Charles was a sergeant from 7 November 1863 until 15 March 1865. [PEN].netlify .app/ pension-file .html - Compiled Military Service Record, Charles Gannsen, Battery B., rolls 327 and 762; Declaration of Recruit, Volunteer Enlistment and surgeon’s form for examining a recruit, 2 Dec 1862, St. Louis; Missouri Adjutant General, descriptive rolls, 2nd Regiment Artillery, images 20, 179, 183; Historical Register of National Homes for Disabled Volunteer Soldiers, p. 2072, no. 2072. [CSR] [ENL62] [AGO] [DAN]
- Roster of the Department of Missouri, Grand Army of the Republic, p. 39. [GAR]
- U.S. War Department, The War of the Rebellion, vol. 22, pp. 344, 417 and 765. [STATIONS]
- St. Marcus Evangelical Church (St. Louis, Missouri), registers of marriages, baptisms, confirmations, deaths and burials, 1850–1903, FamilySearch image group number (IGN) 007769184: marriage 1866, p. 135, entry 676/4 (image 381); confirmation classes 1881–1894; burial register no. 3393 (1903); imaged, FamilySearch (https://
www : accessed 29 October 2026); citing “St. Marcus Evangelical Church records, 1847-1981,” microfilm of originals at the Eden Archives in St. Louis, Missouri. [STM].familysearch .org/ ark:/ 61903/ 3:1:3Q9M-CSMZ-1NJC - St. Marcus Evangelical Church (St. Louis, Missouri), marriage registers 1850–1880, images 324–449, image 331, image 355, p. 98, image 55; St. Marcus Evangelical Church (St. Louis, Missouri), registers of marriages. Confirmation registers, classes 1881–1894, images 240–258. [WALL] [STM]
- 1880 U.S. census, first enumeration (5 June 1880), St. Louis, SD 1, ED 40, p. 14, dwelling 77, family 135, “Ganser, Charles,” 36. In June 1880 he was living at 1621 Jackson Street. 1880 U.S. census, second enumeration (9 November 1880), St. Louis, ED 137, p. 4D (penned), dwelling 15, image 204; 1890 U.S. census, Special Schedule of Surviving Soldiers, ED 55; 1900 U.S. census, St. Louis, Ward 9, enumeration district 129, sheet 5-B, line 96. In June 1900 he was boarding at 170 Sidney Street. [C1880J] [C1880] [VET90] [C1900]
- St. Louis city directories (Kennedy’s, Edwards’, Gould’s), 1860–1904, read year by year in the images under every spelling of the surname and under Johnson. [DIRS]
- St. Louis Recorder of Deeds, Book 382 pp. 392–393 (lease to “Charles Konson,” 1869); Book 391 pp. 478–479 (deed of trust of Rose Carstens, by her trustee Henry Carstens, 18 Oct 1869, on the barroom outfit he later bought); Book 400 pp. 170–171 (deed of trust, Charles Kanson to Louis A. Raum, a barroom outfit from Jacob Orlemann, 18 Jan 1870); imaged, FamilySearch, “Deed records of the city and county of St. Louis, Mo., 1804-1901”: Book 382, image group number (IGN) 008099520, image 239 (https://
www : accessed 4 November 2026); Book 391, IGN 008115150, image 599 (https://.familysearch .org/ search/ film/ 008099520 ?i =238 www : accessed 4 November 2026); Book 400, IGN 008115155, image 125 (https://.familysearch .org/ ark:/ 61903/ 3:1:3Q9M-CSV7-Y828 www : accessed 4 November 2026); St. Louis Court of Criminal Correction, naturalization record, 21 Oct 1868, vol. 10, p. 126; index card imaged, “Missouri, County Naturalization Records, 1830-1985,” FamilySearch, image group number (IGN) 007787201, image 4805 (https://.familysearch .org/ ark:/ 61903/ 3:1:3Q9M-CSV7-KSZK-2 www : accessed 9 November 2026); St. Louis Register of Voters, Ward 9, 1896-series book, line 22, image 191; City of St. Louis, certificate of death, burial permit no. 10619 (register no. 4166), image 1009. [DEEDS] [NAT] [VOTE] [DC03].familysearch .org/ ark:/ 61903/ 3:1:3Q9M-CSMB-S9W3-G - Westliche Post (St. Louis, Missouri), 23 Aug 1867, p. 2 (picnic notice, “Charles Gannsen”; image 641135228), and 15 Jun 1868, p. 2 (advertisement, “Charles Gansen”; image 640596759); Anzeiger des Westens (St. Louis, Missouri); St. Louis Globe-Democrat; St. Louis Post-Dispatch; Missouri Republican (St. Louis, Missouri); all read on ; the two Westliche Post pages imaged, (https://
www , https://.newspapers .com/ image/ 641135228/ www : accessed 7 November 2026); Westliche Post (St. Louis, Missouri), 19 Dec 1903, p. 8, image 640863401. [NEWS] [WP03].newspapers .com/ image/ 640596759/ - Steven Rowan, “The German Press in St. Louis and Missouri in the Nineteenth Century: The Establishment of a Tradition,” Papers of the Bibliographical Society of America 99, no. 3 (2005): 459–67, pp. 460–61, https://
www . JSTOR text layer, read 1 Nov 2026. [ROWAN05].jstor .org/ stable/ 24295912 - Steven Rowan, “The German Press in St. Louis and Missouri in the Nineteenth Century,” Papers of the Bibliographical Society of America 99, no. 3 (2005), p. 467. [ROWAN05]
- Landeskirchliches Archiv der Evangelischen Kirche von Kurhessen-Waldeck, Kassel, parish registers of Kirchenkreis Hofgeismar, read in the images on Archion (https://
www : accessed 11 November 2026). [ARCH].archion .de - Parish search coverage, 57 places of Kreis Hofgeismar. [COVER]
- The Trendelburg researcher, transcriptions and family charts (Stammtafeln) for families of Deisel. [Trendelburg researcher's charts, 2025]
- Hessisches Staatsarchiv Marburg (HStAM), Best. 180 Hofgeismar, Nr. 128, Nr. 841; “Periodische Übersichten über die stattgefundenen Auswanderungen” [Periodic summaries of the emigrations that took place], vol. 4 (1858–1861), image 207, fol. 204. [HSTAM] [H3243]
- Landesgeschichtliches Informationssystem Hessen (LAGIS), Hessische Auswanderer, Nr. 128. [LAGIS59]
- Eberhardt’s Allgemeiner Polizei-Anzeiger, vols. 36–64 (1853–1866), Bayerische Staatsbibliothek, Münchener Digitalisierungszentrum (https://
www , e.g. bsb10767130, vol. 36, 1853), also listed in the Deutsches Zeitungsportal (accessed 30 October 2026); every full-text hit for “Hofgeismar,” “Deisel” and “Trendelburg” read (71 hits for Hofgeismar on 8 Oct 2026; the portal counted 68 issues on 30 Oct 2026). [POLIZEI].digitale-sammlungen .de - Wochenblatt für die Provinz Fulda (Fulda), 1858, Nr. 8, p. 82, image 96. [WBFULDA58]
- Wochenblatt für die Provinz Oberhessen (Marburg), 1858, Nr. 39, p. 723, image 730. [WBOBERH58]
- Place-keyed and name-keyed searches of the Hamburg passenger lists (Ancestry collection 1068); Immigrant Ships Transcribers Guild, vol. 16. [SHIPS] [ISTG]
- Klaus Kunze, Ortssippenbücher (HeiKun-Verlag) for Helmarshausen (1521–1950); Database of births, deaths and marriages for Gottsbüren and the surrounding parishes (“Gottsbüren und Umgebung”). [OSB] [Gottsbüren database (Dornette)]
- Martina Lüdicke, „Verschwunden ist der Strand…“ Studien zur Auswanderung nach Amerika aus Kurhessen in den Jahren 1831–1866, dargestellt am Beispiel des Dorfes Deisel (Kreis Hofgeismar) [“The shore has vanished…”: studies of emigration to America from Kurhessen in the years 1831–1866, shown by the example of the village of Deisel (Kreis Hofgeismar)] Die Geschichte unserer Heimat 14 (Hofgeismar: Verein für hessische Geschichte und Landeskunde, Zweigverein Hofgeismar, 1994). Read in full in a copy borrowed through interlibrary loan, 5 Nov 2026. Pages 32 (the count), 60 (the letters), 69–70 (the sources) and 86–89 (the list). [LUED]
- Johannes Köster, letters to his wife and children at Kassel, 6 Nov 1858 – 2 Dec 1862, pp. 1–55. [KOESTER]
- FamilyTreeDNA Big Y-700 test of a direct male-line descendant of Charles; Autosomal tests of a descendant of Charles at FamilyTreeDNA, AncestryDNA and MyHeritage. [BIGY] [AUTO]
- Thomas W. Jones, “Writing about, Documenting, and Publishing DNA Test Results,” in Advanced Genetic Genealogy: Techniques and Case Studies, ed. Debbie Parker Wayne (Cushing, Tex.: Wayne Research, 2019), 277–302, p. 291 (standards and codes of ethics require permission to publish information, including genetic data, about living people), p. 292 (without permission, names, locations, close relatives and DNA kit numbers are withheld), p. 299 (sources readers cannot consult, such as a personal match list behind one password, carry little value; transfer or anonymized data where possible). Print copy, Harold B. Lee Library, Brigham Young University, via interlibrary loan; pages read in photographs, 8 November 2026; place, publisher and year from the title page and its verso. Page 299. [JONES19]
- Maarten H. D. Larmuseau, Pieter van den Berg, Sofie Claerhout, no. 23. [LARMUSEAU19]
- Ev. Kirchengemeinde Deisel, “Trauungen 1830–1901” [Marriages 1830–1901], image 1305, p. 12, no. 29. [AND1833]
- Autosomal tests of a descendant of Charles: FamilyTreeDNA Family Finder match list and segment export (7 Jun 2026; match list searched again 10 and 18 Oct 2026; close-relative link 18 Oct 2026; compared with a close relative of the tested descendant, 21 Oct 2026); AncestryDNA ethnicity estimate, match list, ThruLines and parental side assignment (28 Jun, 10 Oct, 18 Oct and 21 Oct 2026); MyHeritage DNA, Ethnicity Estimate v2.5, match list, AutoClusters and chromosome browser (7 Jun, 18–19 Jul, 10 Oct, 18 Oct and 22 Oct 2026); FamilyTreeDNA paternal assignment by Family Matching, 22 Oct 2026, reviewed again 23 Oct 2026; father’s-side matches searched for the surnames of all proposed candidate families, 22 Oct 2026; AncestryDNA father’s-side matches with trees in the villages of Kreis Hofgeismar checked for shared matches among themselves, 28 Oct 2026 (no group found); the close relative’s AncestryDNA match list searched for the sisters’ families and villages, 29 Oct 2026; MyHeritage chromosome browser, segment positions (build 37) and triangulation, 28 and 31 Oct 2026; AncestryDNA and MyHeritage match trees searched for the Hofedietz line, 2–3 Nov 2026; MyHeritage match sizes and segment counts read again 6 Nov 2026. Living matches are identified by code and not named. [AUTO53]
- Blaine T. Bettinger, The Shared cM Project, version 4.0 (March 2020), averages and ranges of shared centimorgans by relationship, as presented in DNA Painter’s Shared cM Project tool (dnapainter
.com/ ), read 29 Oct 2026, 31 Oct 2026 and 3 Nov 2026. The tool’s relationship probabilities come from AncestryDNA’s simulations, digitized by Leah Larkin, as given by DNA Painter; the simulations themselves were not read. [SCP453]tools/ sharedcmv4 - Patricia Lee Hobbs, “Correlating Documentary and DNA Evidence to Identify an Unknown Ancestor,” in Advanced Genetic Genealogy, p. 252, p. 253. [HOBBS19]
- Autosomal tests of a descendant of Charles. [AUTO53]
- Blaine T. Bettinger, The Shared cM Project, version 4.0 (March 2020), averages and ranges of shared centimorgans by relationship, as presented in DNA Painter’s Shared cM Project tool (dnapainter
.com/ ), read 29 Oct 2026, 31 Oct 2026 and 3 Nov 2026. The tool’s relationship probabilities come from AncestryDNA’s simulations, digitized by Leah Larkin, as given by DNA Painter; the simulations themselves were not read. [SCP453]tools/ sharedcmv4 - Ev. Kirchengemeinde Trendelburg, Taufen 1830–1863 (Archion register 220108), Nro 58–333 (1833–1841) read from the body leaves. [TBAP]
- Ev. Kirchengemeinde Trendelburg, Taufen 1830–1863 (Archion register 220108), Nro 58–333 (1833–1841) read from the body leaves. [TBAP]
- Evangelische Kirchengemeinde Deisel (Deisel, Kreis Hofgeismar, Kurhessen), “Taufen 1830–1861” [Baptisms 1830–1861] (Archion register 218842), image 51. [PERSON]
- 1870 U.S. census, St. Louis city and county, https://
www . [C1870].familysearch .org/ search/ collection/ 1438024 - Compiled Military Service Record, Charles Gannsen, Battery B., rolls 327 and 762. [CSR]
- Board for Certification of Genealogists, Genealogy Standards, 2nd ed. rev. (Nashville, Tenn.: Ancestry, 2021). E-book edition (2nd ed., rev.), chapter 1, “The Genealogical Proof Standard.” [BCG]
- Elizabeth Shown Mills, Evidence Explained. Page 11 (the Evidence Analysis Process Map). [MILLS]
- Declaration of Recruit, Volunteer Enlistment and surgeon’s form for examining a recruit, 2 Dec 1862, St. Louis; 1900 U.S. census, St. Louis, Ward 9, enumeration district 129, sheet 5-B, line 96; St. Marcus Evangelical Church (St. Louis, Missouri), burial register, pp. 240–241, no. 3393, image 943. [ENL62] [C1900] [BUR03]
- The surname record: twenty-eight attestations of Charles’s surname in fourteen spellings; St. Marcus Evangelical Church (St. Louis, Missouri), burial register, pp. 240–241, no. 3393, image 943; City of St. Louis, certificate of death, burial permit no. 10619 (register no. 4166), image 1009. [SURN] [BUR03] [DC03]
- Thomas W. Jones, Mastering Genealogical Proof, NGS Special Topics Series, no. 107, chap. 5, pp. 77–78. [JONESMGP]
- The certainty wording used in this book; see “How to Read This Book.” [CW]
- Leah Larkin and Jonny Perl, “What Are the Odds?” (WATO), version 2, DNA Painter (https://
dnapainter : accessed 8 November 2026), a tool that compares hypothetical family trees against shared-cM probabilities simulated by AncestryDNA, with odds calculated by Andrew Millard’s method; DNA Painter, “What Are the Odds? FAQ” (https://.com/ tools/ wato dnapainter : accessed 31 October 2026), on relative scores, on scores that “cannot provide proof on its own,” and on endogamy, which “may significantly affect the scores.” No odds threshold for accepting a hypothesis is given; the 100-to-1 threshold is the author’s. [WATO].com/ help/ wato-faq - Blaine T. Bettinger and Debbie Parker Wayne, Genetic Genealogy in Practice, pp. 13 and 18, pp. 84 and 87. [BETTWAYNE]
- Autosomal tests of a descendant of Charles at FamilyTreeDNA, AncestryDNA and MyHeritage; Blaine T. Bettinger, The Family Tree Guide to DNA Testing and Genetic Genealogy. [AUTO] [BETT]
- Blaine T. Bettinger and Debbie Parker Wayne, Genetic Genealogy in Practice, pp. 13 and 18, pp. 73–74. [BETTWAYNE]
- Autosomal tests of a descendant of Charles at FamilyTreeDNA, AncestryDNA and MyHeritage. [AUTO]
- Leah Larkin and Jonny Perl, “What Are the Odds?” (WATO), version 2. [WATO]
- Blaine T. Bettinger and Debbie Parker Wayne, Genetic Genealogy in Practice, pp. 13 and 18. [BETTWAYNE]
- European Network of Forensic Science Institutes, p. 17, pp. 7 and 22. Forensic scientists are asked to do the same: they list the findings a test could give, with their weight, before the examination. [ENFSI15]
- Brian A. Nosek, Charles R. Ebersole, Alexander C. DeHaven and David T. Mellor, “The Preregistration Revolution,” Proceedings of the National Academy of Sciences 115, no. 11 (13 March 2018): 2600–2606, https://
doi (abstract, p. 2600: the research questions and analysis plan defined “before observing the research outcomes”; section “Challenge 1: Changes to Procedure During Study Administration”: changes and deviations reported transparently); PubMed Central, PMC5856500 (https://.org/ 10 .1073/ pnas .1708274114 pmc : accessed 10 November 2026). Fixing the plan first is what researchers call preregistration. [NOSEK18].ncbi .nlm .nih .gov/ articles/ PMC5856500/ - Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset; Leah Larkin and Jonny Perl, “What Are the Odds?” (WATO), version 2; Jed Johnson, comp., “DNA decision plan, fixed 10 November 2026,” blind setting. [PREREG] [WATO] [DNAPLAN]
- Colin Aitken, Paul Roberts and Graham Jackson, Fundamentals of Probability and Statistical Evidence in Criminal Proceedings: Guidance for Judges, Lawyers, Forensic Scientists and Expert Witnesses, Communicating and Interpreting Statistical Evidence in the Administration of Criminal Justice, Practitioner Guide no. 1 (Royal Statistical Society, n.d.; dated November 2010 on the Society’s “Statistics law guides” page, https://
rss ), para. 2.28, p. 46 (posterior odds equal the likelihood ratio times the prior odds), and paras. 2.32–2.33, p. 50 (independent events and the product rule) (https://.org .uk/ news-publication/ publications/ law-guides/ rss : accessed 10 November 2026). Paragraph 2.28, p. 46. The odds after a piece of evidence are the odds before it multiplied by its likelihood ratio. Multiplying rests on an assumption: that the DNA result is independent of the record clues, given who Charles was. Nothing in the records can change what a cousin’s DNA shows, so the assumption is reasonable, but it is still an assumption (paragraphs 2.32–2.33, p. 50). [RSS10].org .uk/ RSS/ media/ News-and-publications/ Publications/ Reports %20and %20guides/ rss -fundamentals -probability -statistical -evidence .pdf - Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- European Network of Forensic Science Institutes, p. 17. [ENFSI15]
- Maarten H. D. Larmuseau, Joost Vanoverbeke, Anneleen Van Geystelen, no. 1772; Maarten H. D. Larmuseau, Koen Matthijs and Tom Wenseleers, no. 5; Maarten H. D. Larmuseau, Pieter van den Berg, Sofie Claerhout, no. 23; Blaine T. Bettinger, The Shared cM Project, version 4.0 (March 2020), averages and ranges of shared centimorgans by relationship, as presented in DNA Painter’s Shared cM Project tool (dnapainter
.com/ , accessed 6 November 2026). [LARMUSEAU13] [LARMUSEAU16] [LARMUSEAU19] [SCP4]tools/ sharedcmv4 - Jed Johnson, comp., “Branch facts for I-Y38863 (10 Nov 2026),” in “The data behind Charles Gannsen and His Family” (dataset compiled for this book; see the dataset’s README): tester counts and countries on I-Y38863 and the branches above it, from the FamilyTreeDNA Discover data files (tree version 2026.11.1; https://
discover ) and YFull YTree v14.07.00 (https://.familytreedna .com/ y-dna/ I-Y38863/ story www ), read 10 November 2026; living testers given as counts and countries only. [BRANCHF].yfull .com/ tree/ I-Y38863/ - YFull, “What Is YFull’s Age Estimation Methodology?,” FAQ, last updated 20 February 2021 (https://
www : accessed 10 November 2026): reliable SNPs counted in the combBED regions (8,473,821 base pairs) and an assumed rate of one SNP in 144.41 years, with an assumed age of 60 for living testers; the method is based on Adamov, Guryanov, Korzhavin, Tagankin and Urasin, “Defining a New Rate Constant for Y-Chromosome SNPs Based on Full Sequencing Data” (2015), as given in the FAQ (the paper itself not read); Jed Johnson, comp., “Branch facts for I-Y38863 (10 Nov 2026),” in “The data behind Charles Gannsen and His Family”. [YFULLFAQ] [BRANCHF].yfull .com/ faq/ what -yfulls -age -estimation -methodology/ - Genetic Genealogy Standards Committee, “Genetic Genealogy Standards” (10 January 2015), standards 15, “Limitations of Y-DNA Testing” (p. 2), 19, “Interpretation of DNA Test Results,” and 20, “DNA as Part of Genealogical Proof” (p. 3), PDF linked from the committee’s site, https://
www (https://.geneticgenealogystandards .com/ www : accessed 10 November 2026). Standards 15, 19 and 20. [GGS15].thegeneticgenealogist .com/ wp-content/ uploads/ 2015/ 01/ Genetic -Genealogy -Standards .pdf - European Network of Forensic Science Institutes, p. 17. [ENFSI15]
- Jed Johnson, comp., “DNA decision plan, fixed 10 November 2026,” blind setting; Jed Johnson, comp., “Branch facts for I-Y38863 (10 Nov 2026),” in “The data behind Charles Gannsen and His Family”. [DNAPLAN] [BRANCHF]
- Jed Johnson, comp., “DNA decision plan, fixed 10 November 2026,” blind setting. [DNAPLAN]
- Jed Johnson, comp., “DNA decision plan, fixed 10 November 2026,” blind setting. [DNAPLAN]
- Board for Certification of Genealogists, Genealogy Standards, 2nd ed. rev. (Nashville, Tenn.: Ancestry, 2021), standards 9, “Planned research” (p. 11); 17, “Extent” (p. 14); 38, “Source preference” (pp. 23–24); and 58–61, “Genealogical Proofs” (pp. 33–35). Standards 9, “Planned research,” and 17, “Extent.” [BCG21]
- Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41. [LEDGER]
- Working positions of the research, October 2026. [DEC]
- Declaration of Recruit, Volunteer Enlistment and surgeon’s form for examining a recruit, 2 Dec 1862, St. Louis; 1900 U.S. census, St. Louis, Ward 9, enumeration district 129, sheet 5-B, line 96. [ENL62] [C1900]
- Declaration of Recruit, Volunteer Enlistment and surgeon’s form for examining a recruit, 2 Dec 1862, St. Louis; St. Marcus Evangelical Church (St. Louis, Missouri), marriage register 1866, p. 135, entry 676/4, image 381. [ENL62] [MAR66]
- Parish search coverage, 57 places of Kreis Hofgeismar. [COVER]
- St. Marcus Evangelical Church (St. Louis, Missouri), registers of marriages. [STM]
- Hessisches Staatsarchiv Marburg (HStAM), Best. 180 Hofgeismar, Nr. 128, Nr. 841. [HSTAM]
- The share is a place in the order of work if Charles is one of the open boys, not the chance that Karl Ludwig was Charles.
- Board for Certification of Genealogists, Genealogy Standards, 2nd ed. rev. (Nashville, Tenn.: Ancestry, 2021), standards 9, “Planned research” (p. 11). Standard 38, “Source preference.” [BCG21]
- Parish search coverage, 57 places of Kreis Hofgeismar; Baptism registers of the Ring 1 parishes read twice, 1833–41. [COVER] [R1READS]
- Database of births, deaths and marriages for Gottsbüren and the surrounding parishes (“Gottsbüren und Umgebung”); Klaus Kunze, Ortssippenbücher (HeiKun-Verlag) for Helmarshausen (1521–1950); Klaus Kunze, comp., “Gesamtpersonenregister der Ortssippenbücher: Oberes Weserbergland,” a combined name index of nineteen village family books edited by Klaus Kunze, André Ausmeyer and Rolf Nowak; it gives name, birth or death year and family number (https://
www : searched 24 and 26 October 2026; accessed 3 November 2026). [Gottsbüren database (Dornette)] [OSB] [KUNZEGPR].klauskunze .com/ familienforschung/ gesamtregister .htm - Working positions of the research, October 2026. [DEC]
- Elizabeth Shown Mills, “QuickLesson 11: Identity Problems & the FAN Principle,” Evidence Explained: Historical Analysis, Citation & Source Usage (https://
www : posted 26 August 2012; accessed 3 November 2026). [MILLSFAN].evidenceexplained .com/ content/ quicklesson -11 -identity -problems -fan -principle - Re-examination, 10 Oct 2026, of the eleven key eliminations against the three-identifier image standard. [AUDIT]
- Board for Certification of Genealogists, Genealogy Standards, 2nd ed. rev. (Nashville, Tenn.: Ancestry, 2021), standards 9, “Planned research” (p. 11). [BCG21]
- Board for Certification of Genealogists, Genealogy Standards, 2nd ed. rev. (Nashville, Tenn.: Ancestry, 2021), standards 9, “Planned research” (p. 11). [BCG21]
- Thomas W. Jones, Mastering Genealogical Proof, NGS Special Topics Series, no. 107, chap. 5. [JONESMGP]
- Elizabeth Shown Mills, Evidence Explained; Elizabeth Shown Mills, “QuickLesson 11: Identity Problems & the FAN Principle,” Evidence Explained: Historical Analysis. [MILLS] [MILLSFAN]
- The Chicago Manual of Style, 18th ed. (Chicago: University of Chicago Press, 2024), especially chap. 3, “Illustrations and Tables.” [CMOS18]
- Plain Language Action and Information Network (PLAIN), Federal Plain Language Guidelines, March 2011, rev. 1, May 2011 (Washington, D.C.: PLAIN, 2011); archived copy of the Guidelines’ table of contents on , CyberCemetery, University of North Texas Libraries (https://
cybercemetery : captured 6 October 2012; accessed 8 November 2026). [PLAIN11].unt .edu/ oilspill/ 20121006185804/ http:// www .plainlanguage .gov/ howto/ guidelines/ FederalPLGuidelines/ TOC .cfm - Ivan P. Fellegi and Alan B. Sunter, “A Theory for Record Linkage,” Journal of the American Statistical Association 64, no. 328 (December 1969): 1183–1210, https://
doi ; E. A. Wrigley, ed., Identifying People in the Past (London: Edward Arnold, 1973). [FELLEGI69] [WRIGLEY73].org/ 10 .1080/ 01621459 .1969 .10501049 - European Network of Forensic Science Institutes, p. 17. [ENFSI15]
- Anthony O’Hagan, “Expert Knowledge Elicitation: Subjective but Scientific,” The American Statistician 73, supplement 1 (2019): 69–81, https://
doi ; and Anthony O’Hagan and Jeremy Oakley, SHELF: The Sheffield Elicitation Framework, version 4 (2019) (https://.org/ 10 .1080/ 00031305 .2018 .1518265 tonyohagan : accessed 3 November 2026). [SHELF19].co .uk/ shelf/ - Ruth Gilbert, Rosemary Lafferty, Gareth Hagger-Johnson, Katie Harron, Li-Chun Zhang, Peter Smith, Chris Dibben and Harvey Goldstein, “GUILD: GUidance for Information about Linking Data Sets,” Journal of Public Health 40, no. 1 (March 2018): 191–198, https://
doi (accessed 3 November 2026). [GUILD18].org/ 10 .1093/ pubmed/ fdx037 - Gary S. Collins, Johannes B. Reitsma, Douglas G. Altman and Karel G. M. Moons, “Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD): The TRIPOD Statement,” BMJ 350 (7 January 2015): g7594, https://
doi (accessed 3 November 2026). [TRIPOD15].org/ 10 .1136/ bmj .g7594 - Thomas A. Lang and Douglas G. Altman, “Basic Statistical Reporting for Articles Published in Biomedical Journals: The ‘Statistical Analyses and Methods in the Published Literature’ or the SAMPL Guidelines,” International Journal of Nursing Studies 52, no. 1 (January 2015): 5–9, https://
doi (accessed 3 November 2026); Ronald L. Wasserstein and Nicole A. Lazar, “The ASA Statement on p-Values: Context, Process, and Purpose,” The American Statistician 70, no. 2 (2016): 129–133, https://.org/ 10 .1016/ j .ijnurstu .2014 .09 .006 doi (accessed 3 November 2026). [SAMPL15] [ASA16].org/ 10 .1080/ 00031305 .2016 .1154108 - Yoav Benjamini and Yosef Hochberg, “Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing,” Journal of the Royal Statistical Society, Series B (Methodological) 57, no. 1 (1995): 289–300, https://
doi (accessed 3 November 2026). [BH95].org/ 10 .1111/ j .2517-6161 .1995 .tb02031 .x - Mark D. Wilkinson, Michel Dumontier, IJsbrand Jan Aalbersberg and others, “The FAIR Guiding Principles for Scientific Data Management and Stewardship,” Scientific Data 3 (15 March 2016): article 160018, https://
doi (accessed 3 November 2026). [FAIR16].org/ 10 .1038/ sdata .2016 .18 - Carlo Ginzburg, “Microhistory: Two or Three Things That I Know about It,” trans. John and Anne C. Tedeschi, Critical Inquiry 20, no. 1 (1993): 10–35, at 23–24, https://
www . Read through UVU JSTOR, 4 Nov 2026; checked in the page images. [GINZBURG93].jstor .org/ stable/ 1343946 - William S. Cleveland, The Elements of Graphing Data, rev. ed. (Summit, N.J.: Hobart Press, 1994); William S. Cleveland and Robert McGill, “Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods,” Journal of the American Statistical Association 79, no. 387 (September 1984): 531–554, https://
doi ; Edward R. Tufte, The Visual Display of Quantitative Information, 2nd ed. (Cheshire, Conn.: Graphics Press, 2001). [CLEVE94] [CLEVMCG84] [TUFTE01].org/ 10 .1080/ 01621459 .1984 .10478080 - Ivan P. Fellegi and Alan B. Sunter. [FELLEGI69]
- E. A. Wrigley, ed., Identifying People in the Past (London: Edward Arnold, 1973). [WRIGLEY73]
- Anthony O’Hagan, “Expert Knowledge Elicitation: Subjective but Scientific,” The American Statistician 73. [SHELF19]
- 1900 U.S. census, St. Louis, Ward 9, enumeration district 129, sheet 5-B, line 96; St. Marcus Evangelical Church (St. Louis, Missouri), burial register, pp. 240–241, no. 3393, image 943. [C1900] [BUR03]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; M405 roll 327 (roll image 1560), Gannsen file: Battery muster-out roll, St. Louis, image 228767518, p. 2. [ODDS] [CSRMO]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg, Deisel and Friedrichsfeld, 2–3 November 2026: 3 full-method cases and 12 ranking cases (plus 2 sensitivity cases), answers sealed until scoring; rerun 5 November 2026 with the current weighing (the author’s and the pooled judgments) on the list of 426 boys, and enlarged to 17 ranking cases (plus 4 sensitivity cases), the 5 added cases set out from American records but not blind; exact Clopper–Pearson intervals, calibration of the shares against the true men, a check of the calibration method on the frozen Ring 1 record-linkage model of 8 November 2026 with 16 grooms from the Dalhausen, Eberschütz and Lamerden, and Beverungen marriage registers, each read twice and matched to his baptism (the calibrated weighing of the open boys has had no such test), AUC and point-biserial correlation with cluster-bootstrap ranges (test design, case files, scores and scoring scripts in the dataset; see the dataset’s README). [VALID]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41. [LEDGER]
- Jed Johnson, comp., “Boys baptized at Deisel, Trendelburg and Friedrichsfeld, 1833–1841,” dataset compiled for this book. [DSFATES]
- Jed Johnson, comp., “Boys baptized at Deisel, Trendelburg and Friedrichsfeld, 1833–1841,” dataset compiled for this book. The 422 are those of the 426 boys baptized whose birth month is recorded. The forty (9 percent) and the thirty-eight (9 percent) are these shares of the 422. They are the base rates against which the ranking weighs the two dates. [DSFATES]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., “Boys baptized at Deisel, Trendelburg and Friedrichsfeld, 1833–1841,” dataset compiled for this book. About one boy in nine was born out of wedlock, a little more often in Trendelburg and Friedrichsfeld than in Deisel: 48 of the 426 boys baptized in the three parishes. A span made this way holds the true share nineteen times in twenty, given how few cases there are; Friedrichsfeld’s is wide because it baptized so few boys. Twenty of the 48 died young, 42 percent against 35 percent of all the boys. [DSFATES]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [PREREG]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys, and the run that added the court-judgment clue, 11–12 November 2026: “Model v2 HOLD test, pooled rounds 2 and 3: result, with a bookkeeping correction” (18 sealed cases, with the approved re-score on the Ring 1 pools); “Model 3d, round-4 HOLD scoring” (15 sealed cases, each change scored once, after a failed plumbing check was fixed); “Adopting model B23: T refit, tail check and Karl Ludwig’s new share,” 12 November 2026; “Model 3e, round-5 HOLD scoring” (15 sealed cases, each change scored once, after a preflight fix; five cases flagged before scoring); “Adopting B23r + 1r: tail check and Karl Ludwig’s new share,” 12 November 2026; for the court-judgment clue, “Amendment 3f run: the Kassel court judgments in the model,” 12 November 2026; and, for the Konze ties weighed by one rule, “Amendment 3g run: Karl Ludwig 68.1%,” 12 November 2026 (with the frozen settings, codings, scores and scripts, in the research dataset); and, for the year of leaving weighed by the measured error of census immigration years, “Amendment 3j: the 1900 census year as a measured spread,” 16 November 2026, with its two audits; and, for Karl Ludwig’s departure window, “Amendment 3k: the land act of 1857 in the model,” 16 November 2026 (same dataset). The search of all 390 boys is “Court judgments and summonses for all 390 boys,” 12 November 2026 (in the research dataset). The rules are Amendments 3c, 3d, 3e, 3f, 3g, 3j and 3k to the preregistration (see PREREG). [ODDS] [MODTEST]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [ODDS] [PREREG]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys; Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [MODTEST] [PREREG]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- European Network of Forensic Science Institutes, p. 17. [ENFSI15]
- Elizabeth Shown Mills, Evidence Explained. Section 1.1, p. 20. [MILLS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Staatsarchiv Hamburg, Auswanderungsamt, 373-7 I, VIII A 1 Band 010, p. 788; Passenger list of the Donau, arrived New York 15 November 1856, line 255, roll 168. [ODDS] [HAMB56] [NYARR56]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Anthony O’Hagan, “Expert Knowledge Elicitation: Subjective but Scientific,” The American Statistician 73. [ODDS] [SHELF19]
- Jed Johnson, comp., sensitivity of the open-boys weighing to the reading of the 1862 middle initial (L, no weight, C, and C standing for any later name), 12 November 2026, pooled and author’s judgments, 20,000 runs each, with 80 and 95 percent uncertainty intervals (script and output in the research dataset); Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [INITSENS] [ODDS]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [PREREG]
- Jed Johnson, comp., “Amendment 3h §4: the middle-initial count (run 12–13 November 2026),” with its briefs, coding sheets, joining script and output; and “Middle-initial count: the ten-case check, as made by the delegated checker (12 November 2026),” with the record of that deviation from the signed amendment; all in the research dataset, with the count’s result for each case. American middle initials from records of each man’s lifetime only; baptismal names from the sealed keys and the ledger of the boys; the ten check cases read in the record images. [INITCOUNT]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Ev. Kirchengemeinde Deisel, “Konfirmationen 1831–1926” [Confirmations 1831–1926], image 931, no. 12, image 941, no. 14. [DCONF]
- Landesgeschichtliches Informationssystem Hessen (LAGIS), Hessische Auswanderer, Nr. 128. [LAGIS59]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Hessisches Staatsarchiv Marburg (HStAM), Protokolle, II Deisel 3 Bd. 3c, pp. 1019–1020, images 229 and 230. [PROT57]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [ODDS] [PREREG]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41; Database of births, deaths and marriages for Gottsbüren and the surrounding parishes (“Gottsbüren und Umgebung”). [LEDGER] [Gottsbüren database (Dornette)]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. The shares use the judgments of the author and two blind reviewers pooled and the result calibrated. If it is not certain that he is one of them, every share is smaller (see the table above). The interval is the 10th to 90th percentile of 20,000 runs; in each, one of the three sets of judgments was picked at random and each judgment drawn between its low and high value. The clues are treated as independent, except Carl and the parish, which are weighed together because Carls were much rarer at Deisel. Base rates come from the 276 boys of the three home parishes who lived to fourteen (birth dates: all 426 baptisms). n = 27 open boys. [ODDS]
- Thomas W. Jones, Mastering Genealogical Proof, NGS Special Topics Series, no. 107, chap. 5; Board for Certification of Genealogists, Genealogy Standards, 2nd ed. rev. (Nashville, Tenn.: Ancestry, 2021), standards 9, “Planned research” (p. 11). Standards 58–61, under “Genealogical Proofs.” [JONESMGP] [BCG21]
- Jed Johnson, comp., “Kurrent reading test, 13–14 November 2026”: blind readings of 98 Deisel, Sielen, Stammen and Langenthal register entries (Archion registers 218851, 218842, 218845, 218854, 218824, 219703, 220033, 220045, 220069, 220078) by two AI models, scored field by field against readings settled earlier (two agreeing reads, or a disagreement settled by Kunze’s family books or another record) and corrected by an image check; with Archion’s and Transkribus’s machine text on the disputed fields. Score sheets, script and corrections in the research dataset. [READTEST]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. The twenty-seven are in the order of the weighing; the three marked * are from the confirmation classes of the neighboring parishes and are outside it. Each cell gives the weight in bits, after calibration; the key gives the signs. One bit doubles or halves the odds. The share is each boy’s chance, pooled and calibrated, if Charles is one of the open boys, with its 80 percent uncertainty interval in parentheses; a share under 1 percent is given as “under 1.” The clues are treated as independent, except Carl and the parish, which are weighed together. [ODDS]
- Jed Johnson, comp., “Kurrent reading test, 13–14 November 2026”. [READTEST]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- C. J. Clopper and E. S. Pearson, “The Use of Confidence or Fiducial Limits Illustrated in the Case of the Binomial,” Biometrika 26, no. 4 (1934): 404–413, https://
doi . [CLOPPER34].org/ 10 .1093/ biomet/ 26 .4 .404 - Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- C. J. Clopper and E. S. Pearson. [CLOPPER34]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg; C. J. Clopper and E. S. Pearson. [VALID] [CLOPPER34]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [PREREG]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [MODTEST]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [MODTEST]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [PREREG]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [MODTEST]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [MODTEST]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys; Wochenblatt für die Provinz Oberhessen (Marburg), 1858, Nr. 39, p. 723, image 730. [MODTEST] [WBOBERH58]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [PREREG]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [MODTEST]
- Jed Johnson, comp., the Konze ties of the twenty-seven open boys searched to the same depth, 12 November 2026. [KTIES]
- Jed Johnson, comp., the Konze ties of the twenty-seven open boys searched to the same depth, 12 November 2026. [KTIES]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [PREREG]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [ODDS]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [ODDS] [MODTEST]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset; Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [PREREG] [MODTEST]
- Jed Johnson, comp., “How firm is ‘immigrated 1859’? A test of census immigration years against known departures,” 16 November 2026: the 1900 (or 1910) census immigration year of 31 men of Deisel, Trendelburg, Friedrichsfeld and the next ring of parishes who arrived in 1840–66, each scored against a departure or arrival year from another record (a passenger list, a release in the emigration index, a register remark), with the tables of every pair (in the research dataset). [CENS59]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [ODDS] [PREREG]
- Hessisches Staatsarchiv Marburg (HStAM), Protokolle, II Deisel 3 Bd. 3c, pp. 1019–1020, images 229 and 230. [PROT57]
- Ev. Kirchengemeinde Deisel, “Konfirmationen 1831–1926” [Confirmations 1831–1926], image 931, no. 12, image 941, no. 14. [DCONF]
- Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg; Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [ODDS] [PREREG]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- Jed Johnson, comp., sealed tests of proposed changes to the weighing of the open boys. [MODTEST]
- Jed Johnson, comp., blind test of the search method on known emigrants of Trendelburg. [VALID]
- Karl Ludwig Köhler, baptism, 19 March 1839, p. 38, no. 175. [KAHLER]
- Jacob Cohen, “A Coefficient of Agreement for Nominal Scales,” Educational and Psychological Measurement 20, no. 1 (April 1960): 37–46, https://
doi ; J. Richard Landis and Gary G. Koch, “The Measurement of Observer Agreement for Categorical Data,” Biometrics 33, no. 1 (March 1977): 159–174, at 165 (the benchmarks, 0.81–1.00 “Almost Perfect”), https://.org/ 10 .1177/ 001316446002000104 doi . [COHEN60] [LANDIS77].org/ 10 .2307/ 2529310 - Baptism registers of the Ring 1 parishes read twice, 1833–41. [R1READS]
- Jed Johnson, comp., data-quality table of the Charles Gannsen project, 3 November 2026: agreement, Cohen’s kappa with bootstrap ranges and first-reading accuracy with Wilson ranges, computed from the second-reading files of the Ring 1 parishes and the logged sample reads of the Ring 0 ledger (file and script in the dataset; see the dataset’s README). [DQ]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset; Jed Johnson, comp., likelihood-ratio model of the twenty-seven open boys of Trendelburg. [PREREG] [ODDS]
- The certainty wording used in this book. [CW]
- Working positions of the research, October 2026. [DEC]
- FamilyTreeDNA Big Y-700 test of a direct male-line descendant of Charles; Autosomal tests of a descendant of Charles at FamilyTreeDNA, AncestryDNA and MyHeritage. [BIGY] [AUTO]
- Jed Johnson, comp., “What would change the conclusion: criteria fixed in advance, 3 November 2026,” and “Amendment 1 to the preregistration of 3 Nov 2026,” both in the docs folder of “The data behind Charles Gannsen and His Family” (dataset compiled for this book, version of 3 November 2026; see the dataset’s README), written before the results of the lying-in, Ring 1, Marburg, Duisburg and DNA searches it names; their SHA-256 checksums are in CHECKSUMS.sha256. “Amendment 2 to the preregistration of 3 Nov 2026,” dated 11 November 2026 and made before any DNA result, in the docs folder of the same dataset, changes the Big Y rule. “Amendment 3 to the preregistration of 3 Nov 2026: the 1855–62 conscript summonses,” dated 11 November 2026, adds the summons clue; its weights were set blind, but it was added after the result was seen. “Amendment 3b to the preregistration of 3 Nov 2026: validation of the candidate model,” of the same date, sets how the weighing is tested on known men. “Amendment 3c to the preregistration of 3 Nov 2026: two proposed changes to the candidate model, and how they will be judged,” dated 11 November 2026, sets how changes to the weighing are tested. “Amendment 3d to the preregistration of 3 Nov 2026: three proposed improvements to the candidate model, and how they will be judged,” of the same date, proposes three more changes. Each was signed before the sealed cases it governs were scored. “Amendment 3e to the preregistration of 3 Nov 2026: a revised departure clue and a refit of the model on 73 known men, and how they will be judged” sets the second round. It is dated and signed 12 November 2026, before its sealed cases were scored. “Amendment 3f to the preregistration of 3 Nov 2026: the Kassel court judgments of 1855–62,” dated and signed 12 November 2026 before the clue was run, adds the judgments as new evidence. Their weights were set blind on 11 November 2026. “Amendment 3g to the preregistration of 3 Nov 2026: Konze ties coded for every open boy at the same depth,” dated 12 November 2026 and signed that day before its share was computed, scores every open boy’s Konze ties by one rule, priced by a new blind setter. When its no-Charles check failed at the old setting for Charles’s own clues, that setting was refitted by the rule of Amendment 3c, as 3g allows, with the author’s written approval. “Amendment 3h to the preregistration of 3 Nov 2026: the final round, and the freeze,” dated and signed 12 November 2026 before the middle-initial count was made, fixes the form of the weighing, fixes the setting for Charles’s own clues at 2.21, sets the range printed for the chance that Charles is among the open boys, and sets the rule for counting middle initials and for using the count. Its ten-case check of the count, which it gives to the author, was delegated by him; the record of that deviation is kept beside it. “Amendment 3j to the preregistration of 3 Nov 2026: the 1900 census year as a measured spread,” dated 16 November 2026, was written after the author lifted the freeze of 3h for one clue, the year of leaving; its main form was named before any share under it was computed, but its shares were computed before it was signed. “Amendment 3k to the preregistration of 3 Nov 2026: the land act of 1857 in the model,” of the same date, was written after the author lifted the freeze again, for one record: Karl Ludwig’s departure window, taken from the act that reports him in America by 5 December 1857; its share too was computed before it was signed. All ten amendments are in the docs folder of the same dataset. [PREREG]
- Thomas W. Jones, Mastering Genealogical Proof, NGS Special Topics Series, no. 107, chap. 5. [JONESMGP]
- Parish search coverage, 57 places of Kreis Hofgeismar. [COVER]
- Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41; Trendelburg baptisms Jul 1830 – 1832 (Archion 220099, image 1267–1273; 220108, image 1378–1395), read in the images; Deisel baptisms 1830–1832 (218824, image 1746–1749 and 1783; 218842, image 4–22), read entry by entry; confirmation classes 1842–1844 at Trendelburg (220102, image 2254–2260) and 1843–1845 at Deisel (218836, image 923–927). Carl Heinrich Steitz of Deisel (born 19 Nov 1830, confirmed 1845) was Henry Steitz of Charleston, South Carolina, who swore in 1880 that he was born at “Deissel, Hessen” on 19 Nov 1830 (U.S. passport application no. 16913, NARA M1372; FamilySearch ark 1:1:QV5Y-7K38). Read 11–12 Oct 2026; imaged, Archion (https://
www ), by the register numbers and images given; citing Landeskirchliches Archiv der EKKW, Kassel. [LEDGER] [B1830].archion .de - Ev. Kirchengemeinde Trendelburg, Taufen 1830–1863 (Archion register 220108), Nro 58–333 (1833–1841) read from the body leaves. The totals for those three years are 37, 35 and 26 children, and they agree with the entries by number and by sex. [TBAP]
- Landesgeschichtliches Informationssystem Hessen (LAGIS), Hessische Auswanderer, Nr. 128; Jed Johnson, comp., “Boys baptized at Deisel, Trendelburg and Friedrichsfeld, 1833–1841,” dataset compiled for this book. LAGIS is the Hessian state’s online historical information system. The bars count 70 boys: 62 of the 67 traced overseas and 8 open boys with a departure on record. The other five of the 67 are on neither list. The hatched bar is a capture–recapture estimate (Chapman) with its 95 percent confidence interval: a range made this way holds the true number of unseen emigrants nineteen times in twenty; the open bar is the 19 open boys with no departure on record. The method’s assumptions are only partly met: the two kinds of record were not searched independently. [LAGIS59] [DSFATES]
- Parish search coverage, 57 places of Kreis Hofgeismar; Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41. [COVER] [LEDGER]
- Parish search coverage, 57 places of Kreis Hofgeismar. [COVER]
- Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41. [LEDGER]
- Landesgeschichtliches Informationssystem Hessen (LAGIS), Hessische Auswanderer, Nr. 128. [LAGIS59]
- Hessisches Staatsarchiv Marburg, replies to Jed Johnson, 26–28 Oct 2026, Nr. 1546. [HSTAM28]
- Wochenblatt für die Provinz Fulda (Fulda), 1858, Nr. 8, p. 82, image 96; Wochenblatt für die Provinz Oberhessen (Marburg), 1858, Nr. 39, p. 723, image 730; Deutscher Reichsanzeiger und Preußischer Staatsanzeiger, Nr. 54, 1 Mar 1900, images 14–15. [WBFULDA58] [WBOBERH58] [RA1900]
- Place-keyed and name-keyed searches of the Hamburg passenger lists (Ancestry collection 1068). [SHIPS]
- St. Marcus Evangelical Church (St. Louis, Missouri), marriage registers 1850–1880, images 324–449, image 331, image 355, p. 98, image 55. [WALL]
- St. Louis Court of Criminal Correction, naturalization, 21 Oct 1868, p. 126. [NAT68]
- Declaration of Recruit, Volunteer Enlistment and surgeon’s form for examining a recruit, 2 Dec 1862, St. Louis; Compiled Military Service Record, Charles Gannsen, Battery B., rolls 327 and 762; Charles Gannson (Private, Battery B, 2nd Missouri Light Artillery, Civil War), pension application no. Inv. 749,432, certificate no. 684,507; Case Files of Approved Pension Applications of Veterans Who Served in the Army and Navy Mainly in the Civil War and the War with Spain, 1861–1934; Civil War and Later Pension Files; Record Group 15: Records of the Department of Veterans Affairs; National Archives, Washington, D.C.; digital copy of the file supplied to the author by the National Archives, imaged at https://
charles-research : accessed 30 October 2026. Mail-order copies ordered by the author, first a partial file, order F21-314123835E, submitted 14 February 2017 and delivered electronically, and then the complete file, order F41-523680269E, submitted 8 September 2019 and sent by mail. Charles was a sergeant from 7 November 1863 until 15 March 1865. [ENL62] [CSR] [PEN].netlify .app/ pension-file .html - FamilyTreeDNA Big Y-700 test of a direct male-line descendant of Charles; Autosomal tests of a descendant of Charles at FamilyTreeDNA, AncestryDNA and MyHeritage. [BIGY] [AUTO]
- Professional researcher engaged by the author to work at the Hessisches Staatsarchiv Marburg, emails to Jed Johnson, 13 Nov 2026, Nr. 1546, 180, Nr. 128, Nr. 1998. [MRES1113]
- Landeskirchliches Archiv Kassel (Az. 43.1) to Jed Johnson, email, 30 Oct 2026. [LKAK1030]
- Universitätsbibliothek Kassel to Jed Johnson, email, 2 Nov 2026; Wochenblatt für die Provinz Niederhessen (Kassel), Nr. 20, 16 Feb 1858. [UBK1102] [WB1858]
- Professional researcher engaged by the author to work at the Hessisches Staatsarchiv Marburg, emails to Jed Johnson, 13 Nov 2026, Nr. 1546, 180, Nr. 128, Nr. 1998. [MRES1113]
- Parish search coverage, 57 places of Kreis Hofgeismar; Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41. [COVER] [LEDGER]
- Baptism registers of the Ring 1 parishes read twice, 1833–41. [R1READS]
- Database of births, deaths and marriages for Gottsbüren and the surrounding parishes (“Gottsbüren und Umgebung”). [Gottsbüren database (Dornette)]
- Parish search coverage, 57 places of Kreis Hofgeismar. [COVER]
- Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41. [LEDGER]
- Jed Johnson, comp., measurement of how far men of the generation born about 1825–45 moved. [RADIUS]
- Jed Johnson, comp., measurement of how far men of the generation born about 1825–45 moved; Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41. [RADIUS] [LEDGER]
- Jed Johnson, comp., measurement of how far men of the generation born about 1825–45 moved. [RADIUS]
- Jed Johnson, comp., measurement of how far men of the generation born about 1825–45 moved, 8 November 2026: the birthplaces of the grooms born in 1825–45 in the marriage registers of Trendelburg and Friedrichsfeld, 1855–75 (Archion 220120 and 218959), used only where two independent readings of the register agree; the later German places of the boys born at Trendelburg, Deisel and Friedrichsfeld in 1833–41, from the ledger (index-only entries left out); great-circle distances from place coordinates, percentiles with 95 percent percentile-bootstrap intervals (5,000 resamples, seed 20261108), and lower and upper bounds for the places without coordinates; of the 104 grooms of the first test, 16 had a birthplace inferred (from residence, the parents or the register’s own parish) rather than stated, and on stated birthplaces alone 81 of the 82 that could be placed were born within 15 km of the parish where they married; the script, its output and the data for each man are in the research dataset. [RADIUS]
- Confirmation registers read entry by entry for every class 1845–1858, image 427–438; Confirmation registers of Ehrsten (Archion 218917), Fürstenwald (218983), Wilhelmshausen (220324), Schachten (Kirchenbuch 219991), Burguffeln (Kirchenbuch 218749), Ersen (218947), Mönchehof (Kirchenbuch 219850), Eberschütz (218884), Niedermeiser (219874) and Veckerhagen (220207), the classes of 1856 or 1857 to 1860, read entry by entry with the pastors’ totals, about 280 boys, Oct 2026 (at Ersen no class was held in 1857). No Konze, Kontze, Conze, Konzen, Gans or Gansen; no Carl born on or near 26 January; no note of emigration. The one Trendelburg-born boy, Heinrich August Carl Kreßling, born 29 June 1844 and confirmed at Burguffeln on 11 April 1858 (image 1444), is of another family. LKA Kassel via Archion; imaged, Archion (https://
www : accessed 2 November 2026), by the register numbers and images given; Confirmation registers read entry by entry, with the pastors’ year totals, for every class 1845–1858: Hümme (Archion 219529), Gottsbüren (219046), Bad Karlshafen Lutheran (218674) and Reformed (218707); Confirmation registers read entry by entry, with the pastors’ year totals: Zwergen (Archion register 220357), Udenhausen (220141) and Vaake (220171), classes 1845–1858; Hohenkirchen (219412), 1845–1857; Obermeiser (219907), 1845–1856; Westuffeln (220300), Carlsdorf (218806), Calden (218776), Griemelsheim (Kirchenbuch 219145) and Knickhagen (Kirchenbuch 219643), classes 1845–1858; Ehrsten (218917), Fürstenwald (218983), Wilhelmshausen (220324) and Schachten (Kirchenbuch 219991), 1845–1856; Burguffeln (Kirchenbuch 218749), 1845–1857; Ersen (218947) and Mönchehof (Kirchenbuch 219850), 1845–1855; Wilhelmsthal (Kirchenbuch 220345, images 390–393), classes 1845–1858, in which no boy was confirmed. Carl Osterberg: Schachten, class of 1847, image 511, no. 1; baptism Kirchenbuch 219991, image 389, 1833 no. 17 (born at Schachten). The classes of these outer parishes from 1856 or 1857 to 1860 were read later, without result (Oct 2026): Ehrsten, Fürstenwald, Wilhelmshausen, Schachten, Burguffeln, Ersen, Mönchehof, Eberschütz, Niedermeiser and Veckerhagen; imaged, Archion (https://.archion .de www : accessed 2 November 2026) > Hessen > Landeskirchliches Archiv der Evangelischen Kirche von Kurhessen-Waldeck > Kirchenkreis Hofgeismar, by the register numbers given; citing Landeskirchliches Archiv der Evangelischen Kirche von Kurhessen-Waldeck, Kassel; Evangelische Kirchengemeinde Grebenstein (Grebenstein, Kreis Hofgeismar, Kurhessen), “Taufen 1840–1854” [Baptisms 1840–1854] (Archion register 219109), image 45. [CONF401] [CONF60] [CONFRING] [Confirmation registers, 1845–58] [KONTZEGR].archion .de - Baptism registers 1833–1841 of eighteen parishes of Kirchenkreis Hofgeismar. [KBAP]
- Baptism registers of the Ring 1 parishes read twice, 1833–41; Jed Johnson, comp., searches of American records for the boys of the Ring 1 parishes. The twenty Ring 1 parishes have twenty-two baptism registers. Of the 1,357 who died in Germany, nearly all died as children. Of the 44 who emigrated, 22 are known only from American records. Of the 1,194 still open, 9 are probably traced. The open boys were also checked against the burials to about 1855: at a second reading for Stammen, Langenthal, Ostheim, Liebenau, Hofgeismar, Herstelle, Beverungen and Dalhausen, and for the rest at a single reading or a partial second reading. [R1READS] [R1US]
- Jed Johnson, comp., measurement of how far men of the generation born about 1825–45 moved. [RADIUS]
- Database of births, deaths and marriages for Gottsbüren and the surrounding parishes (“Gottsbüren und Umgebung”). [Gottsbüren database (Dornette)]
- 1860 U.S. census, St. Louis city and county. [C1860]
- 1870 U.S. census, St. Louis city and county. [C1870]
- Place-keyed and name-keyed searches of the Hamburg passenger lists (Ancestry collection 1068). [SHIPS]
- Jed Johnson, comp., “The data behind Charles Gannsen and His Family,” version of 3 November 2026; Forward trace of every boy baptized at Trendelburg, Deisel and Friedrichsfeld born 1833–41; Parish search coverage, 57 places of Kreis Hofgeismar. [DSREADME] [LEDGER] [COVER]
- Thomas W. Jones, “Writing about, Documenting, and Publishing DNA Test Results,” in Advanced Genetic Genealogy, p. 291. [JONES19]
- Thomas W. Jones, “Writing about, Documenting, and Publishing DNA Test Results,” in Advanced Genetic Genealogy, p. 292. [JONES19]
Supplements
Records, images and data on this website that go with this chapter.








